A Survey on Context-Aware Multi-Agent Systems: Techniques, Challenges and Future Directions

TL;DR

This paper addresses the need for a coherent framework at the intersection of context-aware systems and multi-agent systems by proposing a general CA-MAS process (Sense-Learn-Reason-Predict-Act) and surveying state-of-the-art techniques across domains like autonomous navigation, disaster relief, and IoT. It offers a cohesive framework linking CAS and MAS through awareness, learning, reasoning, and coordination, and discusses challenges such as noisy context sharing, security, and consensus uncertainty, along with future directions. The contributions include a taxonomy of CA-MAS components, architectural guidance, and identified research gaps to advance robust context-driven multi-agent collaboration. Overall, the work provides a foundation for designing adaptive, context-aware autonomous systems capable of operating in dynamic and uncertain environments.

Abstract

Research interest in autonomous agents is on the rise as an emerging topic. The notable achievements of Large Language Models (LLMs) have demonstrated the considerable potential to attain human-like intelligence in autonomous agents. However, the challenge lies in enabling these agents to learn, reason, and navigate uncertainties in dynamic environments. Context awareness emerges as a pivotal element in fortifying multi-agent systems when dealing with dynamic situations. Despite existing research focusing on both context-aware systems and multi-agent systems, there is a lack of comprehensive surveys outlining techniques for integrating context-aware systems with multi-agent systems. To address this gap, this survey provides a comprehensive overview of state-of-the-art context-aware multi-agent systems. First, we outline the properties of both context-aware systems and multi-agent systems that facilitate integration between these systems. Subsequently, we propose a general process for context-aware systems, with each phase of the process encompassing diverse approaches drawn from various application domains such as collision avoidance in autonomous driving, disaster relief management, utility management, supply chain management, human-AI interaction, and others. Finally, we discuss the existing challenges of context-aware multi-agent systems and provide future research directions in this field.

Figures (7)

• Figure 1: An Overview of Context-Aware Multi-Agent Systems
• Figure 2: Ten organizational structures of multi-agent systems. A blue circle symbolizes an agent, with its role depending on the organizational structure. For instance, in the federation structure (d), agents fulfill two distinct roles: (1) normal agents depicted as blue circles, and (2) federated agents shown as outlined circles. Similarly, in the market structure (h), there are two roles: buyers represented by blue circles and sellers by outlined circles. In the matrix structure (i), managers are represented as blue circles while workers are shown as outlined circles.
• Figure 3: Coordination strategies between agents. A blue circle represents an agent.
• Figure 5: An general process of context-aware systems. A blue circle represents sensed information from an agent. During the context utilization stage, one possible scenario is where Agent A observes communication between Agents B and C. In this situation, Agent A observes and understands the communication between Agents B and C by retrieving their contexts and relationships. Similarly, Agents B and C understand each other's information by retrieving each other's contexts.
• Figure 7: Architectures of context-aware systems. A blue circle represents the information generated by diverse sensors.