External Human-Machine Interfaces for Autonomous Vehicles to Pedestrian Communication

posted on 29 Nov 2021 by jeannie

A research study to understand the socio-behavioural aspect of pedestrians when interacting with autonomous vehicles (AVs).

Brief

This research aims to understand how pedestrians interact with autonomous vehicles (AVs). The findings from conducted socio-behavioral user studies will contribute to the creation of an External Human-Machine Interface Simulator System that is used to evaluate safe pedestrian-AV interactions. The envisaged outcome is to develop interfaces, guidelines, or recommendations for pedestrians to safely interact with AVs from Singapore’s context.

Objectives

The project comprises of two main objectives:

The first is to develop a set of recommendations to educate human pedestrians on how to interact with AVs, and a set of behavioral guidelines for AVs, based on socio-behavioral user studies.

Conducting pedestrian socio-behavioral user studies and design workshops will help us understand how pedestrians use the road alongside with human drivers under controlled road conditions. These investigations provide knowledge of how people interact and communicate their intentions with other drivers and identify fundamental human instincts and behavioral characteristics that can be leveraged upon. The findings from these investigations will provide valuable insights for the guidelines and recommendations developed.

The second objective is to design and implement a software simulator system for external vehicular interface, behaviors, designs, specifications, or guidelines for AVs.

The development of a software External Human-Machine Interface simulation system has the potential to be used to convey potential AV behaviors through use of external interfaces. This is especially useful since some of the critical events like near-miss accidents cannot be intentionally duplicated under real-world conditions for testing. User studies with the software deployment will be conducted to understand the potential of the guidelines and designs. Using the results of the evaluations, design guidelines and specifications for External Human-Machine Interfaces for Autonomous Vehicles can be further proposed.

Singapore Context

AV researches were done in Europe, normally SAE 0, 3, and 5 vehicles are simulated and certain types of crossing scenarios. Few exists in Singapore or even Asian context. Hence, this project opens an avenue to contribute in a similar research area for Asia.

Members

SIT

Undergraduate Research Assistants

Acknowledgements

This work is funded by:

  1. Research & Development of Methods for a Holistic Evaluation of AV Safety under Tropical Urban Conditions, LTA (UMGC-L010)