Skip to menu Skip to content Skip to footer
Dr Stephany Berrio Perez
Dr

Stephany Berrio Perez

Email: 

Overview

Background

Stephany Berrio is a robotics researcher specialising in perception, mapping, and safety for autonomous vehicles, working at the intersection of machine learning and real-world deployment. Her research helps self-driving systems understand complex, changing urban environments and operate safely alongside people. Her current focus areas include:

  • Multimodal perception and sensor fusion — developing algorithms that fuse camera, LiDAR, and radar data for robust scene understanding, semantic occupancy prediction, HD mapping, domain adaptation, and anomaly detection in intelligent transportation systems.
  • Long-term mapping and localisation — enabling vehicles to maintain accurate position estimates as cities change over time.
  • Open datasets and benchmarks — building publicly available multimodal datasets for anomaly segmentation, V2X collaboration, and driving in challenging conditions such as rain and rural roads.

Stephany completed her PhD at the University of Sydney's Australian Centre for Field Robotics on scene understanding and map maintenance for autonomous vehicles.

Availability

Dr Stephany Berrio Perez is:
Available for supervision

Research impacts

Stephany's research moves autonomous vehicle technology out of the lab and onto real roads — shaping government policy, supporting industry, and making driving safer. Stephanycontributed to the Connected and Automated Vehicle (CAV) trials led by the Australian Centre for Robotics and supported by Transport for NSW (running since 2020). These on-road trials in complex urban settings generated evidence used to inform future government policy on how automated vehicles can be safely integrated into the road network.

Open resources for a global research community. She co-created several publicly available datasets and benchmarks that lower the barrier for safety research worldwide, including:

  • STU (Spotting the Unexpected) — the first public 3D LiDAR dataset for road anomaly segmentation (CVPR 2025), helping researchers test how vehicles detect unexpected hazards
  • Mixed Signals — a diverse point cloud dataset for vehicle-to-infrastructure (V2X) collaboration (ICCV 2025), developed with international partners including Cornell University
  • Panoptic-CUDAL — a rural Australian dataset captured in rainy conditions (ITSC 2025), addressing the under-studied challenge of driving beyond cities.

Together, these activities support safer roads, stronger local capability in a globally competitive industry, and evidence-based regulation of a technology that will reshape everyday transport.

Works

Search Professor Stephany Berrio Perez’s works on UQ eSpace

44 works between 2012 and 2026

41 - 44 of 44 works

2018

Conference Publication

Octree map based on sparse point cloud and heuristic probability distribution for labeled images

Berrio, Julie Stephany, Zhou, Wei, Ward, James, Worrall, Stewart and Nebot, Eduardo (2018). Octree map based on sparse point cloud and heuristic probability distribution for labeled images. 25th IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid Spain, Oct 01-05, 2018. NEW YORK: IEEE. doi: 10.1109/iros.2018.8594024

Octree map based on sparse point cloud and heuristic probability distribution for labeled images

2018

Journal Article

Digital image processing applied on static sign language recognition system/Diseño de un Sistema de Reconocimiento de Gestos No Móviles mediante el Procesamiento Digital de Imágenes

Villa, Betsy, Valencia, Valeria and Berrio, Julie (2018). Digital image processing applied on static sign language recognition system/Diseño de un Sistema de Reconocimiento de Gestos No Móviles mediante el Procesamiento Digital de Imágenes. Prospectiva, 16 (2), 41-48. doi: 10.15665/rp.v16i2.1488

Digital image processing applied on static sign language recognition system/Diseño de un Sistema de Reconocimiento de Gestos No Móviles mediante el Procesamiento Digital de Imágenes

2014

Conference Publication

A "FLIPPED CLASSROOM" FOR MOBILE ROBOTICS TEACHING

Berrio Perez, Julie Stephany (2014). A "FLIPPED CLASSROOM" FOR MOBILE ROBOTICS TEACHING. 8th International Technology, Education and Development Conference (INTED), Valencia Spain, Mar 10-12, 2014. VALENICA: IATED-INT ASSOC TECHNOLOGY EDUCATION A& DEVELOPMENT.

A "FLIPPED CLASSROOM" FOR MOBILE ROBOTICS TEACHING

2012

Journal Article

Lines Extraction from Laser Scans Integrating Hough Transform, Total Least Squares and Successive Edges Following

Berrio P, J. Stephany, Ordonez, Samith O. and Caicedo, Eduardo F. (2012). Lines Extraction from Laser Scans Integrating Hough Transform, Total Least Squares and Successive Edges Following. Ingenieria, 17 (1), 48-59.

Lines Extraction from Laser Scans Integrating Hough Transform, Total Least Squares and Successive Edges Following

Supervision

Availability

Dr Stephany Berrio Perez is:
Available for supervision

Looking for a supervisor? Read our advice on how to choose a supervisor.

Media

Enquiries

For media enquiries about Dr Stephany Berrio Perez's areas of expertise, story ideas and help finding experts, contact our Media team:

communications@uq.edu.au