ST Kinetics Autonomous Bus Trial (15 May 2017 – 14 November 2020)

The Autonomous Bus (AB) needs to have a communication system between the AB computer system and the on-board passengers. Without a driver on board, the AB would need to have additional communication channels with the passengers, for example, to inform passengers the status and navigation plan of autonomous driving, to greet passengers, to alert passengers of danger, and to respond to emergency requests. This research will study the multi-modal, multi-media techniques to support the software module, and to integrate the software with the decision system of the AB. The human-AB communication for on-board passengers focuses on dialogue system for on-board passengers. In a traditional bus, on-board passengers usually move to the front to speak to the bus driver when there is a need. Without a driver on board, the AB would need to have additional communication/interaction channels for the passengers, for example, to greet the passengers, to answer to inquiries specific to the AB, and to respond to emergency requests. This research will develop a passenger-AB speech dialogue system. Through an interactive kiosk or passenger’s own smartphone, passengers can speak to the AB onboard the AB. Passengers can speak to the AB agent via an interactive kiosk and speak to the control centre when there is need. The first phase deployment is an onboard kiosk, which speaks to one passenger at a time. There is an option to extend the AB agent through a smartphone application, through which multiple passengers can speak to the AB at the same time.

Project Duration: 15 May 2017 – 14 November 2020

Funding Source: National Research Foundation, Prime Minister’s Office, Singapore

Acknowledgment: This research work is supported by Urban Mobility Grand Challenge Funding Initiative Project Code UM01/002. Project Title: ST Kinetics Autonomous Bus Trial.

PUBLICATIONS

• Yi Fan Ong, Maulik Madhavi, and Ken Chan, “OPENNLU: Open-Source Web-Interface NLU Toolkit for Development of Conversational Agent”, in Proc. Asia-Pacific Signal and Information Processing Association (APSIPA) Annual Summit and Conference (ASC) 2020, Auckland, New Zealand, December 2020.
• Maulik Madhavi, Tong Zhan, Haizhou Li and Min Yuan, “First Leap Towards Development of Dialogue System for Autonomous Bus”, In. Proc. International Workshop on Spoken Dialogue Systems Technology (IWSDS),  Sicily, Italy, April 2019, pp. 1-6. [link]