Self-organising robot swarms could inspire future collision avoidance systems

Robot swarm paves way for better collision avoidance systems

Open platform low cost robots mimicking behaviour of swarming bees pave the way for self-organised cooperation of nature-inspired machines.

The robots, each costing about £25, are equipped with infrared proximity sensors allowing each robot to communicate with its direct neighbours at a range of 0.5 to 2m.

Each of the robots is about 4cm in diameter in size but capable to travel at a rather fast speed of 35 cm/s. A combination of three short-range sensors and an independent processor enables each individual robot to detect obstacles and move effortlessly across rather large distances while being constantly aware of other robots around.

Developed by researchers from the UK University of Lincoln and Chinese Tsinghua University, the self-organising robotic swarm inspired and named after the Colias butterfly, has been described in the recent issue of the International Journal of Advanced Robotic Systems.

“Colias has been designed as a complete platform with supporting software development tools for robotics education and research,” said Farshad Arvin, from the School of Computer Science, University of Lincoln.

“This concept allows for the coordination of simple physical robots in order to cooperatively perform tasks. The decentralised control of robotic swarms can be achieved by providing well-defined interaction rules for each individual robot.”

Colias has been used in a bio-inspired scenario, showing that it is extremely responsive and convenient for investigation of collective behaviours.

Robotic swarms that take inspiration from nature have become a topic of fascination for robotics researchers, whose aim is to study the autonomous behaviour of large numbers of simple robots in order to find technological solutions to common complex tasks.

Due to the hardware complexities and cost of creating robot hardware platforms, current research in swarm robotics is mostly performed by simulation software. However, the simulation of large numbers of these robots in robotic swarm software applications is often inaccurate due to the poor modelling of external conditions.

The Colias team would now like to equip the robot swarm with a technology based on their earlier discovery which mimics locust vision. Equipped with a specific neuron called the ‘lobula giant movement detector’, the insect is promptly alerted to any object approaching its eyes.

The researchers believe that integrating such systems into machines could pave the way for development of highly accurate vehicle collision sensors, surveillance technology and even aid video game programming.

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them