NUS News

Self-driving scooter unveiled

The vision of a car-lite society and efficient door to door mobility led NUS and the Singapore-MIT Alliance for Research and Technology to embark on a collaborative research project on autonomous, self-driving vehicles. The latest in the series is a self-driving scooter that is meant for use in pedestrian environments, to navigate smaller and narrower pathways where larger vehicles cannot easily move around in.

“Mobility is not only outdoors,” explained Associate Professor Marcelo Ang from NUS Mechanical Engineering, co-investigator of the project. “You also want to be mobile in big complexes, like malls. So the concept of mobility is not only on the roads but the total mobility to get me from point A to point B.”

Target users are people who “cannot, should not, or prefer not to drive”, including the elderly, disabled, or the very young. It is hoped that the scooter will eventually be able to work in tandem with previously launched self-driving car and buggy to provide a seamless journey. In a possible scenario, a person could book the autonomous vehicles using an application or webpage, navigate the scooter or buggy from inside a building complex to along pedestrian walkways to the public roads, and ride the car to get to the final destination. The algorithms used in each of the self-driving vehicles are the same, added Assoc Prof Ang, and can be transported from one to another, reducing the need to reprogramme each new vehicle.

The scooter weighs around 50kg and can travel at up to a speed of 6kph. Sensors are built into the device for localisation purposes — to tell where it is, as well as to detect obstacles from up to 2.5m in the front, and 10cm at the sides. It is able to slow down or stop in response to obstacles.

The team’s approach towards constructing the scooter could be called minimalist, shared Assoc Prof Ang. It was built using off-the-shelf components and less expensive sensors than most other self-driving vehicles, making it more economically viable. High levels of self-driving capability were achieved through software, he said.

Ultimately, this driverless technology will contribute towards a car-lite society. This will not only improve accessibility of mobility to the masses but could also reduce in accidents caused by human error, allow better utilisation of commuting time, and benefit the environment as it can reduce emissions by up to 20 to 50 per cent.

This project, supported by the National Research Foundation under its Campus for Research Excellence and Technological Enterprise programme, is currently in its trial stages. Further testing on usage of the vehicle on the road is in progress.