SUTD’s research on a multi-robots system wins Outstanding Paper Award at IEEE MRS 2019
ORION is a first-of-its-kind multi-robots system made up of a wheeled ground unit, miniature and wall-climbing robots to form a scalable swarming system similar to ant colonies
Researchers from the Singapore University of Technology and Design (SUTD) have developed a strong, flexible and scalable multi-robots system (MRS) that can be used in intelligence, surveillance and reconnaissance (ISR) operations. The interdisciplinary research team received the Outstanding Paper Award for their ground-breaking research at the 2nd International Symposium on Multi-Robot and Multi-Agent Systems 2019 (IEEE MRS 2019) held at Rutgers University in New Jersey, United States of America (refer to image).
The researchers drew inspiration from the collective behaviours of ant colonies, flocks of birds, schools of fish and other natural multi-agent swarming systems to create ORION (refer to video). The MRS system includes a combination of a robust wheeled ground unit as well as miniature and wall-climbing robots, allowing it to be scalable since it is able to introduce new robots into the system even while it is operating.
ORION makes use of wall climbing units to support the swarm of mapping robots to navigate across multiple floors and expand the distributed communication network among the ground robots. The wall-climbing units achieve wall-adhesion by means of dry-adhesive wheel-legs also known as ‘whegs’.
Autonomous single-robots have been known to be good contenders for ISR operations as they can move independently and also perform various tasks in remote or hazardous environments preventing direct human intervention. However, compared to MRS, they are unable to effectively explore dynamic environments.
For instance, buildings can be challenging to explore autonomously due to its intermittent accessibility to some spaces caused by closed doors, difficulty in floor-to-floor transfer through staircases or elevators, and failing to establish a communication channel in order to achieve inter-agent information exchanges across walls and floors.
Attempts have been made to address these challenges separately, but this is the first time that a decentralized and heterogeneous system of custom-built miniature robots for autonomous exploration and mapping is developed to simultaneously address these challenges.
“Our research team is extremely proud to receive this recognition for our innovative work in the field of Multi-Agent Systems. I strongly believe that SUTD’s continued success in this line of research comes from our truly cross-disciplinary approach — integrating elements from network science, biology, robotics and control theory to engineering design,” said Associate Professor Roland Bouffanais, Principal Investigator and lead of the Applied Complexity Group at SUTD.