On the Edge of the Kerb

About the workshop

As autonomous vehicles (AVs) approach everyday deployment, Australia is set to largely import them, given it has no domestic vehicle manufacturing industry. The AVs that arrive here, and the external human-machine interfaces (eHMIs) through which they communicate with pedestrians, will have been designed for other places. Whether that communication is legible and trustworthy for Australian pedestrians is an open question.

Australia has a diverse range of urban, regional and remote environments. Its pedestrians differ in age, mobility, vision, language and familiarity with automated systems. Recurring natural events such as bushfires and floods stretch this variability to an extreme that AVs must respond to. This full-day workshop convenes the HCI community to map how AV-pedestrian communication should adapt across Australian contexts and communities. Working in parallel context streams, participants will build a shared, reusable account of contexts, pedestrian groups and communication needs, and develop eHMI concepts grounded in them.

Why Australia is different

Strong sunlight and glare can wash out display-based cues. Signalised crossings are often far apart in the suburbs, so pedestrians commonly cross wide arterial roads mid-block, where no formal infrastructure exists. School zones change their speed limits by time of day. Shared zones mix pedestrians, cyclists and vehicles at 10 km/h. Regional and remote towns have their own road cultures, speeds and distances, and residents who may never have encountered an AV. Many regional and remote areas are home to First Nations communities, for whom AV rollout raises specific considerations about community life, mobility and involvement in decisions about whether and how AVs operate on Country.

The full range of Australian conditions also includes crisis events. Bushfires, floods, cyclones and heatwaves create conditions where ordinary design assumptions break down, and AVs are increasingly being considered, or repurposed on the fly, for crisis roles such as evacuating people who cannot drive. Design knowledge for how vehicles should communicate with people in crisis does not yet exist. This workshop treats crisis as one stream among several: the extreme end of the contextual variability that eHMIs must respond to.

The questions we are asking

1. How do communication needs vary across Australian environments: CBD intersections, suburban streets and school zones, shared zones, transit interchanges, regional towns?
2. How can AVs communicate effectively with Australia's diverse pedestrian groups, across differences in age, mobility, vision, language and familiarity with the technology?
3. Which established, largely overseas, eHMI approaches transfer to Australian conditions, which need adaptation, and where are new approaches required?
4. What changes at the extreme end, when ordinary conditions are impacted by crisis?

Objectives

1. Map Australian contexts and pedestrian groups relevant to AV rollout, and the communication needs each combination implies, establishing a shared and reusable account.
2. Examine the transferability of established eHMI approaches, identifying which transfer to Australian conditions, which need adaptation, and where new approaches are required.
3. Design eHMI concepts for Australian scenarios across distinct context streams, producing preliminary design directions that inform future research, prototyping and evaluation of context-aware eHMIs.
4. Build an Australian HCI research network around AV-pedestrian communication, connecting researchers whose work spans design, urbanism, transport and crisis contexts, with a shared agenda for future research.

Topics of interest

We welcome participants engaging with any of the following.

• How communication needs vary across Australian environments: CBD intersections, suburban arterials and school zones, tram corridors, shared pedestrian-vehicle zones, transit interchanges, and regional and remote towns.
• Pedestrian diversity and inclusion: age, mobility, vision, language and cultural background, distraction, and familiarity with automated systems.
• Transferability of overseas eHMI conventions to Australian conditions, including climate and glare, road rules and culture, and infrastructure.
• Context-aware and adaptive eHMIs that adjust their communication to different environments, situations and pedestrian groups while balancing legibility, usability, trust and safety.
• Multimodal communication (visual, auditory, haptic) for Australian conditions and accessibility needs.
• Crisis and emergency conditions, such as bushfires and floods, as the extreme end of context variation for AV communication.
• The regulatory, infrastructural and policy context of AV rollout in Australia and its implications for interface design.
• Methods for studying and designing AV-pedestrian communication, including in-situ observation, co-design, simulation and scenario-based design.

International HCI researchers are also invited to participate and to contribute their location-specific perspectives.

Call for position papers

Prospective participants are invited to submit a short position paper, 5 to 8 pages in the CEUR-WS single-column format (CEURART), with references not counted towards the page limit. LaTeX users can use the Overleaf template or download the style files from CEUR-WS. Authors who do not use LaTeX should use the ODT template in LibreOffice (free). CEUR-WS does not provide a Word template and advises against Word, Microsoft 365 and Google Docs, which do not support the required Libertinus fonts. Please use the CEURART template as provided. It is single column by default and CEUR-WS no longer accepts two-column papers, so no changes to the document class are needed. Keep the "Declaration on Generative AI" section required by the template, even if only to state that no generative AI tools were used. Submissions should not be anonymised. Proceedings will be submitted to CEUR-WS.org for online publication.

Papers may take the form of a position or provocation, a case study, work in progress, or a reflection on methods, and may address any of the topics of interest above. Authors are encouraged to indicate the context stream closest to their expertise: Australian cities; rural and remote communities; or crisis and emergency.

Submit your paper as a PDF by email to callum.parker@sydney.edu.au with the subject line "OzCHI 2026 workshop submission". Submissions will be reviewed by the organisers and selected for relevance, quality and the range of viewpoints they bring.

Accepted papers will be published as openly accessible workshop proceedings via CEUR-WS.org, giving participants a citable output. At least one author of each accepted paper must register for the workshop and attend in person.

Key dates

• 9 Oct 2026

  Position papers due (Friday, anywhere on Earth).

• 23 Oct 2026

  Notifications to authors.

• 6 Nov 2026

  Camera-ready papers due.

• 18 Nov 2026

  Lightning-talk slide due.

• 21 Nov 2026

  Workshop day: Saturday, at OzCHI 2026 in Adelaide.

Format and schedule

This is a full-day, in-person workshop designed for active participation rather than a sequence of talks. After a shared framing, most of the day is spent in small groups organised as parallel context streams: Australian cities; rural and remote communities; and crisis and emergency. Each stream is facilitated by one of the organisers, and all streams record their work on a common template covering environments, pedestrian groups and communication needs, so outputs can be consolidated into a single shared account spanning everyday to crisis conditions.

The morning establishes the foundation: short provocations, lightning talks from each participant's position paper, a comparative discussion contrasting previous eHMI research with Australian realities, and a stream mapping activity. The afternoon turns to design: each stream develops communication concepts for its scenarios, with a mid-session rotation so everyday concepts are examined under crisis conditions and vice versa. A closing plenary consolidates cross-cutting principles and a shared research agenda.

Schedule of the full-day workshop, 9:00 to 16:00

Time	Session	Details
9:00	Welcome and introduction	Welcome and organiser intros (20 min)
9:20	Provocations	AV rollout and communication in Australia (45 min)
10:05	Lightning talks	One slide per participant; seeds streams (25 min)
10:30	Coffee break	Stream and team formation (20 min)
10:50	Comparative discussion	eHMI defaults vs Australian realities (30 min)
11:20	Stream mapping activity	Contexts, groups and needs on a common template (70 min)
12:30	Lunch	(60 min)
13:30	Concept design activity	Transfer, adapt or invent; cross-stream rotation (90 min)
15:00	Coffee break	Prepare presentations (15 min)
15:15	Group presentations	Each stream presents analysis and concepts (30 min)
15:45	Synthesis and wrap-up	Shared account, agenda, closing (15 min)
16:00	Close

Who should come

The workshop is aimed at HCI and interaction-design researchers working on autonomous vehicles, human-robot interaction, public-space technologies, accessibility and inclusion, or emergency management, with an interest in how AVs will communicate with people across Australian settings. No prior experience with AVs or eHMIs is required; the mix of perspectives across contexts, communities and conditions is the point.

Organisers

• Callum Parker The University of Sydney — Primary contact

  ARC DECRA Fellow and Lecturer in Interaction Design, leading the Urban Interfaces Lab at the University of Sydney. His DECRA project develops methods for designing and evaluating context-aware AV-pedestrian communication across diverse urban contexts and pedestrian groups.

• JF Joel Fredericks The University of Sydney

  Associate Dean (Student Life) and Senior Lecturer in Design at the University of Sydney's School of Architecture, Design and Planning. His research spans interaction design, community engagement and smart urbanism, investigating how co-design, immersive technologies and AI-supported tools can enable inclusive participation, particularly in crisis resilience and urban futures.

• TT Tram Thi Minh Tran The University of Sydney

  Lecturer in Design at the University of Sydney's School of Architecture, Design and Planning. Her research explores the applications and implications of emerging technologies, with a particular focus on AR/VR and autonomous mobility.

• MH Marius Hoggenmueller The University of Sydney

  Lecturer in Interaction Design at the University of Sydney's School of Architecture, Design and Planning. His research focuses on how emerging technologies such as autonomous systems and robots can be designed to communicate with and enrich the experience of people in public urban spaces, including prototyping and evaluating novel interfaces in realistic city environments.

• MF Marcus Foth Queensland University of Technology

  Professor in Strategic Design at Queensland University of Technology and a Chief Investigator in the ARC Training Centre for Automated Vehicles in Rural and Remote Regions (AVR3). His research spans urban interaction design, media architecture, transport and mobility, participatory planning, and data governance in smart cities. He is a Distinguished Member of the ACM.

• MT Martin Tomitsch University of Technology Sydney

  Professor and Head of the Transdisciplinary School at the University of Technology Sydney. His research sits at the intersection of design, technology and urban systems, exploring how pervasive and situated technologies, from public displays to autonomous systems, shape people's interactions in urban environments.

• SK Sven Krome Independent researcher

  UX and design researcher specialising in automated driving systems and human-vehicle interaction. Over the past decade he has led research at Waymo and Uber ATG that informed the design, evaluation and deployment of production autonomous driving technologies, including external human-machine interfaces and in-vehicle experiences.

• RS Ronald Schroeter Queensland University of Technology

  Professor at Queensland University of Technology, the QUT/Seeing Machines Industry Chair in Empathic Machines, and Co-Director of the ARC Training Centre for Automated Vehicles in Rural and Remote Regions (AVR3). He currently chairs the steering committee of the ACM AutoUI conference.