The Role of Robots in Collective Perception: A Self-Organizing Hierarchy Approach

Researchers at the IRIDIA artificial intelligence research laboratory at the Universit√© Libre de Bruxelles have developed a self-organizing approach to improve swarm autonomy in collective perception. In their paper published in Intelligent Computing, the researchers propose a method where one robot temporarily acts as the “brain” to consolidate information on behalf of the group.

This approach combines aspects of centralized and decentralized control, allowing robots to understand their relative positions within the system and fuse their sensor information. It eliminates the need for a global or static communication network or external references. The researchers also apply centralized methods for multi-sensor fusion to a self-organized system for the first time.

To test their approach, the researchers compared it against three benchmark approaches. The self-organizing hierarchy approach outperformed the others in terms of accuracy, consistency, and reaction time. In an experimental setup, a swarm of simulated drones and ground robots collected spatial data and formed a collective opinion of object density using short-range sensors.

The self-organizing hierarchy approach utilizes a “dynamic ad-hoc hierarchical network.” Robots at different levels of the hierarchy have distinct roles in decision-making processes. The top-level robot is responsible for performing inferences and sending motion instructions, while the middle-level robots manage data transfer and participate in balancing global and local motion goals. The majority of robots at the bottom level perform sample collection and manage local motion.

Future research in this area might focus on advanced inference methods and the robustness of sampling techniques in challenging environmental conditions. Overall, this self-organizing approach holds promise for improving collective perception accuracy in swarm robotics.

Source: Aryo Jamshidpey et al, Reducing Uncertainty in Collective Perception Using Self-Organizing Hierarchy, Intelligent Computing (2023).

Note: The source article was referenced, but the URL was not provided.