Function Emergence in Multirobot Systems Utilizing Intelligent Data Carriers.

Teruo Fujii, Hajime Asama, Daisuke Kurabayashi, Hayato Kaetsu, Isao EndoJournal of the Robotics Society of Japan 1999 · DOI

Abstract

This paper describes the use of the newly developed system named “Intelligent Data Carrier System: IDC System” to execute several tasks by a multirobot system in a so-called “emergent” manner. In order to realize the function emergence in real systems, it is important to consider how to design and embed the system into a real form, rather than theoretical or numerical researches on an emergent behavior itself. Based on the discussion on the design frameworks of an “emergent system”, we try to implement the IDC system along with an actual multirobot system which consists of omnidirectional mobile robots equipped with folk lifts for handling of the IDC units. The applicability of the IDC system is demonstrated through the experiments for two kinds of tasks; 1) object transportation, and 2) patrolling. It can be concluded that the IDC system is effectively used in the tasks appearing in the practical missions of multirobot systems which can be regarded as one of the physical (real) systems that could realize “function emergence”.

Introduction

Coordinating a fleet of robots is a significant challenge in robotics. Traditional approaches typically rely on centralized control, which presents a single point of failure, or global communication, which suffers from network congestion as the system scales. This paper proposes a decentralized alternative: the Intelligent Data Carrier (IDC) System. By embedding "intelligence" directly into the environment via local data tags, the authors demonstrate how complex group behaviors—such as object transportation and corridor patrolling—can emerge from simple local interactions.

Conceptual diagram of the IDC system where robots interact with data carriers placed on objects and walls.

System Architecture

The core of the proposed solution is the Intelligent Data Carrier (IDC), a "smart" tag that stores and processes data locally. Unlike passive RFID tags, an IDC unit contains its own CPU (Z80), 32KB of memory, and a wireless communication unit operating at 90MHz FM.

The system architecture is divided into three functional units:

  1. Communication Unit: Handles wireless data transfer between the IDC and the robot's Reader/Writer.
  2. Data Processing Unit: Manages stored information using a CPU and memory.
  3. Power Unit: Controls power supply, including a sleep mode to conserve energy until a robot is within range (~1 meter).

The physical design of the IDC unit features a unique "mushroom" shape, optimized for stacking and handling by the robots' forklifts. The robots themselves are omnidirectional mobile platforms equipped with infrared sensors for obstacle avoidance.

Methodology and Task Implementation

The IDC system was evaluated through two practical multi-robot missions:

Object Transportation

IDCs are attached to physical objects, storing metadata such as the object's ID, destination, and current status (e.g., "waiting for pickup"). Robots patrolling a set path read these IDCs upon encounter and autonomously decide whether to transport the object based on the stored data.

Patrolling and Exploration

IDCs are placed at corridor junctions to store exploration states. When a robot reaches a junction, it reads the IDC to identify unexplored branches, updates the state to "Searching" to prevent redundant efforts by other robots, and proceeds. This allows for coordinated exploration without direct robot-to-robot communication.

Experimental Results

The system's performance was validated through both numerical simulations and physical experiments. A key technical requirement for decentralized coordination is the spatial limitation of communication to ensure local interaction.

Communication success ratio as a function of distance, showing a reliable 1-meter range.

Experimental data confirmed a 100% communication success ratio at distances up to 125 cm, with a sharp drop-off thereafter. This "local-only" communication prevents the "broadcast storm" problem common in global networks. In the patrolling task, three robots successfully explored a complex layout by "leaving notes" for one another on the junction IDCs, demonstrating effective autonomous coordination.

Discussion and Conclusion

The IDC system illustrates the principle of stigmergy—coordination through environmental modification. By treating the environment as a shared memory, the system becomes inherently robust; if a robot fails, the task state remains preserved in the IDC for the next available agent.

Future work will address power management, potentially using non-contact power sources like electromagnetic induction, and the logistical challenges of deploying high densities of IDCs in large-scale environments. This research demonstrates that "function emergence" in multi-robot systems can be achieved by designing a "smarter" environment rather than relying solely on increased robot complexity.

Summary updated

Abstract

This paper describes the use of the newly developed system named“Intelligent Data Carrier System: IDC System”to execute several tasks by a multirobot system in a so-called“emergent”manner. In order to realize the function emergence in real systems, it is important to consider how to design and embed the system into a real form, rather than theoretical or numerical researches on an emergent behavior itself. Based on the discussion on the design frameworks of an“emergent system”, we try to implement the IDC system along with an actual multirobot system which consists of omnidirectional mobile robots equipped with folk lifts for handling of the IDC units. The applicability of the IDC system is demonstrated through the experiments for two kinds of tasks; 1) object transportation, and 2) patrolling. It can be concluded that the IDC system is effectively used in the tasks appearing in the practical missions of multirobot systems which can be regarded as one of the physical (real) systems that could realize“function emergence”.
BibTeX
@article{fujii1999function,
  author = {Teruo Fujii and Hajime Asama and Daisuke Kurabayashi and Hayato Kaetsu and Isao Endo},
  title = {{Function Emergence in Multirobot Systems Utilizing Intelligent Data Carriers.}},
  journal = {Journal of the Robotics Society of Japan},
  year = {1999},
  doi = {10.7210/jrsj.17.848},
}
Figure