SWILT – Swarm Intelligence Layer to Control Autonomous Agents

SWILT – Swarm Intelligence Layer to Control Autonomous Agents

Explore the frontier of agent-based swarm intelligence research with SWILT, a project dedicated to integrating dynamic scheduling and logistical optimizations in industrial settings through novel swarm intelligence algorithms. Within a three-layered architecture, SWILT aims to dissect and understand the complexities of swarm behavior to foster adaptability and resilience in industrial operations. Each agent’s decision contributes to the emergence of efficient and robust industrial ecosystems.
Partner
Lakeside Labs (Coordinator)
Melanie Schranz, Lakeside Labs B04b, 9020 Klagenfurt
https://www.lakeside-labs.com/

Infineon Technologies Austria AG
Walter Laure, Siemensstraße 2, 9500 Villach
https://www.infineon.com/cms/austria

University of Klagenfurt
Wilfried Elmenreich, Lakeside B10a, 9020 Klagenfurt
https://nes.aau.at/

Novunex GmbH
RTD, Völkermarkter Ring 7/1, 9020 Klagenfurt
Harnessing the Power of CPS Swarms: Unveiling Complexity Through Nature’s Lens in Industrial Innovations
Cyber-physical systems (CPS) seamlessly integrate software with physical components and find application across diverse fields including IoT, smart mobility, smart energy networks, Industry 4.0, and intelligent homes. These systems connect in networks, utilizing a variety of sensors and actuators to bridge actual and digital realms. A single CPS often operates within a greater collective, leveraging the enhanced dynamics, connectivity, and complexity offered by CPS swarms. This requires capabilities such as adaptability, scalability, robustness, and autonomous organization, due to the increased complexity and challenges in managing and programming such swarms. SWILT is dedicated to exploring the utilization of CPS swarms within industrial settings, including the coordination of products, machinery, and equipment. To navigate the intricacies of managing CPS swarms, it looks to the natural world for inspiration, where numerous identical agents collaborate autonomously, following simple local directives.

Project Details

The project started on October 1, 2018, and concluded on September 30, 2021. Spanning from 2018 to 2021, the overall project duration was 36 months. The project concluded with multiple achievements, notably in the realm of knowledge dissemination and media outreach. In terms of academic contributions, the outcomes included an array of publications in both journals and conferences, highlighting the project’s research findings and innovations. This extensive body of work serves as a testament to the project’s scientific rigor and its contribution to the broader academic community.

Media and Press Releases

In addition to the scholarly output, the project also made notable appearance in the media landscape. There were multiple press releases, capturing the attention of news outlets ranging from “Kleine Zeitung” to “Der Standard”. This media coverage helped in improving the project’s visibility and impact, bridging the gap between academic research and the general public. Through these efforts, the project not only marked its presence in the scientific domain but also in the societal consciousness, reflecting a successful outreach strategy that amplified its findings and benefits to a wider audience.
Revolutionizing Logistics and Production: SWILT’s Swarm Intelligence Approach
SWILT explores the application areas within production scheduling and logistics transportation, confronting the limitations of linear optimization which only tackles a fraction of industrial operations. This traditional method struggles with the intricacies and dynamic nature of production, rendering it insufficient for navigating the expansive and complex decision spaces characteristic of NP-hard problems, especially when factoring in human operators in the optimization process. Our accumulated experience underscores that self-organizing algorithms are adept at managing NP-hard challenges, offering robust solutions to intricate issues. The breakthrough with SWILT lies not in crafting another fixed schedule or routing guideline, but in harnessing the principles of swarm intelligence. These principles are distilled into rules that operate on autonomous agents locally, resulting in highly adaptable algorithms sensitive to immediate changes.
Unlike static solutions, these algorithms are dynamic, continually processing data to respond to new conditions. This method fundamentally diverges from conventional techniques by fostering solutions that naturally emerge from the interactions of local agents. An ambitious aspect of SWILT is the development of a library dedicated to housing swarm intelligence algorithms, tailored to specific use cases. This library facilitates systematic experimentation, analysis, and benchmarking through simulations. Additionally, overseeing and synchronizing potentially thousands of autonomous agents within a changeable framework presents its own set of challenges. To address this, SWILT incorporates a novel tri-layered structure. The base layer (L3) consists of the agent models, the intermediate layer (L2) forms the swarm network, supporting inter-swarm communication, and is envisioned to integrate seamlessly with 5G technology to meet the high-speed and low-latency demands of Industry 4.0. The top layer (L1) acts as the central oversight mechanism.
This innovative architectural approach opens up numerous research avenues, including the governance of each layer, the refinement of data hierarchy, and the inclusion of human feedback mechanisms to elevate optimization strategies. The project aims to achieve a Technology Readiness Level (TRL) 3, enabling the detailed simulation of inter-layer communication and the efficacy of swarm intelligence in addressing practical scenarios.

ICT of the Future

ICT of the Future is the funding program of the Austrian Federal Ministry for Transport, Innovation and Technology (BMVIT) for the promotion of challenging technology development and innovation in information and communication technology, interlinked with application fields and societal challenges.

A Comprehensive Perspective

The call aims to contribute to the goal of the federal government to lead Austria from the group of so-called Innovation Followers to the group of Innovation Leaders – to become one of the most innovative countries within the EU.