Autonomous Robot Navigation in Human Populated Environment (Moondeep Chandra Shrestha)

When:
April 21, 2019 @ 1:30 pm – 2:30 pm
2019-04-21T13:30:00+02:00
2019-04-21T14:30:00+02:00
Where:
E-JUST – Building (13) flat no.10.

Autonomous Robot Navigation in Human Populated Environment
Moondeep Chandra Shrestha
Waseda University

Abstract:

For the integration of robots into our everyday lives, the ability to perform autonomous navigation among humans is indispensable. Given the importance, robot navigation has been a widely studied field since the 1980s. However, the focus within the navigation field has shifted from more of motion planning approaches towards ‘human aware navigation’ in the last decade or so. The term ‘human-aware navigation’ emphasizes an integral part wherein humans are not just simple dynamic obstacles but social and cooperating agents. Recent researches have prioritized this notion and integrated this philosophy into their motion planning frameworks. This presentation presents a set of research based on a similar philosophy.

The first idea is about employing robot-arm contact for navigation. Conventional researches, almost in its entirety, always emphasize on safety and therefore, prioritize humans. However, in a congested area this approach will result in a highly inefficient robot motion. Therefore, we explore an unconventional idea wherein a robot tries to achieve a more efficient navigation by influencing an obstructing human to move away by means of contact.
The second idea centers around communicating directional intention while navigating. One reason why humans can cooperate seamlessly while navigating is because of their ability to communicate non-verbally. However, unlike humans, present robots cannot convey its intention through human-like non-verbal communication. This presentation explores various existing modes for communicating directional intent of a robot across different passing scenarios as a means of overcoming the shortcomings of the robot’s non-verbal communication abilities. Specifically, studies have been performed into projection indicators, turn indicators, display indicators, and their combinations with sound, and their effectiveness investigated across different passing scenarios.

Bio:

Moondeep Chandra Shrestha received his B.E. in Mechanical Engineering from Tribhuvan University at Nepal on 2010. He received his M.E. and Ph. D degrees in Mechanical Engineering from Waseda University, Japan on 2013, and 2016, respectively. He started working as a Research Associate from 2016 and became an assistant professor in the Department of Modern Mechanical Engineering at Waseda University on 2017. His research interest includes human-robot interaction and autonomous robot navigation. He is a member of IEEE.