Emerging cooperative autonomous systems: Information for control and estimation

Información del proyecto

MINERVA

Identificador del acuerdo de subvención: 101044629

DOI

10.3030/101044629

Fecha de la firma de la CE 12 Julio 2022

Fecha de inicio 1 Diciembre 2022

Fecha de finalización 30 Noviembre 2028

Financiado con arreglo a

• European Research Council (ERC)

Coste total

€ 1 999 686,00

Aportación de la UE

€ 1 999 686,00

Coordinado por

UNIVERSITY OF CYPRUS
Cyprus

Periodic Reporting for period 1 - MINERVA (Emerging cooperative autonomous systems: Information for control and estimation)

Período documentado: 2022-12-01 hasta 2025-05-31

Resumen del contexto y de los objetivos generales del proyecto

The rapid advancement of autonomous systems has led to their increasing deployment in critical domains such as intelligent transportation, industrial automation, and smart cities. These systems rely on the cooperation of multiple autonomous agents to achieve complex tasks more efficiently and adaptively than traditional automated systems. Such systems are often referred to as Cooperative Autonomous Systems (CASs). The integration of CASs with wireless communication networks enables a new level of flexibility, allowing distributed systems to collaborate dynamically without rigid, pre-programmed interactions.

However, the adoption of CASs in real-world applications faces fundamental scientific and technical challenges, particularly in environments with stringent time-sensitive constraints, dynamic uncertainties, and communication limitations. Current design approaches rely on modular methodologies, where control, estimation, and communication are treated as separate layers. This often results in suboptimal, inefficient, and fragile systems that struggle to adapt to real-world conditions. Additionally, the growing complexity of interconnected systems necessitates a paradigm shift—one that moves beyond incremental improvements toward a holistic co-design framework that integrates these critical components seamlessly.

This project aims to bridge the gap between theoretical advances and practical implementations by developing a fundamental yet realistic framework for real-time control, estimation, and localization in CASs. By rethinking the way these systems interact and operate, the project will establish the foundations for future high-performance, adaptive, and robust cooperative autonomous networks.

Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha

The project has made significant progress in advancing communication-aware control, distributed estimation, and multi-agent coordination in cooperative autonomous systems (CASs). The key results include:

1. Communication-Constrained Control & Estimation:
- Developed a framework to evaluate communication reliability, optimizing the trade-offs between compression rate, latency, and reliability.
- Advanced modulation and constellation design to enhance robustness against noise in networked control systems.

2. Robust Wireless Localization & Information-Theoretic Methods:
- Investigated heuristic and learning-based methods for robust wireless localization.
- Developed a systematic framework for the Importance of Information to handle communication impairments and safety constraints.

3. Scalable Multi-Agent Coordination:
- Applied distributed consensus and gradient methods for multi-agent control and coordination.

4. Novel Methodologies for Uncertain Environments:
- Developed distributionally robust estimation techniques for handling faults, uncertainties, and network impairments.
- Deployed a Hybrid Automatic Repeat reQuest (H-ARQ) protocol for resilient data transmission in multi-agent settings.
- Designed dynamic quantization methods for bandwidth-limited communication while ensuring efficient coordination.
- Introduced a finite-time distributed estimation and control gain methodology, allowing fast, decentralized coordination.

Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha)

Scientific results and potential technological impact:
- Advancing networked control theory by unifying estimation, control, and communication.
- Bridging theory and real-world applications through robust, data-driven multi-agent strategies.
- Enabling resilient decision-making in uncertain, dynamic environments.

Key Needs for Further Uptake and Success:

1. Further Research & Demonstration:
- Large-scale simulation and real-world testing to validate the developed frameworks.
- Extending multi-agent reinforcement learning for highly dynamic, decentralized settings.
- Exploring hybrid AI-driven control strategies to improve robustness under real-world conditions.

2. Access to commercialization
- Industry partnerships to co-develop prototypes for autonomous systems.
- Funding for Proof-of-Concept cases in industrial automation and vehicular networks.