top of page





From Core to Swarm Autonomy - Evolving the Future of Autonomous Mobile Robots

In society, autonomy often refers to the ability of a rational individual to make mature, uncoerced decisions. When it comes to robotics, it refers to the machines that have a high degree of ability to act or perform tasks with its core computing power without human intervention ‒ we call it the “Core Autonomy.”

Figure: An autonomous mobile robot (AMR) is an example of Core Autonomy that usually exhibits intelligence such as mobility, AI, and precise control actions.

Some modern industrial robots can work autonomously if they are restricted to a directly related work environment or task. However, the surrounding environment of an industrial robot is often noisy, chaotic, and does not abide by the same restrictions. As a result, this adds unpredictable variables to the operation of the robot.

Swarm Autonomy is exhibited by connecting the Core Autonomy units (autonomous robots)

together. It was inspired by the discipline of swarm intelligence, which observes the behavior of social insects. Swarm Autonomy was discovered by adding communication capabilities to the behavior of individual robots. This enabled complex group behaviors, such as the ability to autonomously decentralize and collaborate like living things and take collaborative and collective actions through the interaction between robots as well as with the surrounding environment.

Figure: A Swarm Robotic System can control large-scale robot formations in a smart factory.
Figure: A Swarm Robotic System can control large-scale robot formations in a smart factory.

A robotic system based on Swarm Autonomy is called a Swarm Robotic System. These systems can carry out large-scale robot formations in a smart factory. The characteristics of Swarm Autonomy include:

  • Awareness: Combines the local data sensed by the robots in a swarm into a global perspective, allowing the swarm to make collective decisions such as avoiding obstacles or other robots.

  • Solidarity: Allows a swarm of robots to autonomously classify tasks and to allocate an appropriate robot or group to perform a specific task.

  • Dynamic reconfiguration: Allows task replication/replacement by any robot to optimize performance, so that any faults caused by deficiencies from individual robots can be recovered automatically.


bottom of page