Language-Grounded Multi-Agent Planning for Personalized and Fair Participatory Urban Sensing
Published on: arXiv:2603.24014v1
Announcement Type: New
Abstract
Participatory urban sensing leverages human mobility for large-scale urban data collection, yet existing methods typically rely on centralized optimization and assume homogeneous participants, resulting in rigid assignments that overlook personal preferences and heterogeneous urban contexts. We propose MAPUS, an LLM-based multi-agent framework for personalized and fair participatory urban sensing. In our framework, participants are modeled as autonomous agents with individual profiles and schedules, while a coordinator agent performs fairness-aware selection and refines sensing routes through language-based negotiation. Experiments on real-world datasets show that MAPUS achieves competitive sensing coverage while substantially improving participant satisfaction and fairness, promoting more human-centric and sustainable urban sensing systems.
Introduction
As urban environments continue to evolve, the need for efficient data collection has become increasingly crucial. Traditional participatory urban sensing approaches have often relied on centralized systems that do not account for the diversity of participant backgrounds and preferences. This oversight can lead to a lack of engagement and suboptimal data collection outcomes.
The MAPUS Framework
MAPUS, or Multi-Agent Planning for Urban Sensing, addresses these challenges by integrating a language-grounded multi-agent system. Here are some key features:
- Individual Profiles: Each participant is represented as an autonomous agent with unique characteristics and schedules.
- Fairness-Aware Selection: The coordinator agent ensures that the selection process is equitable, considering the varying needs of participants.
- Language-Based Negotiation: Participants can negotiate sensing routes through natural language, enhancing collaboration and satisfaction.
Experimental Validation
The effectiveness of the MAPUS framework has been validated through experiments conducted on real-world datasets. Key findings from these experiments include:
- Competitive Sensing Coverage: MAPUS demonstrated an ability to achieve high levels of data collection efficiency.
- Increased Participant Satisfaction: Participants reported a higher level of satisfaction due to the personalized approach.
- Enhanced Fairness: The framework succeeded in promoting fairness among participants, addressing previous shortcomings in urban sensing tasks.
Conclusion
MAPUS represents a significant advancement in the field of participatory urban sensing. By leveraging a multi-agent framework that prioritizes individual preferences and equitable participation, it fosters a more inclusive and effective data collection environment. The findings suggest that this approach not only enhances data quality but also encourages greater community involvement in urban sensing initiatives.
Future Directions
Further research is required to explore the scalability of the MAPUS framework in various urban contexts and to refine the negotiation protocols used by participants. Additionally, incorporating feedback mechanisms could further enhance participant engagement and data accuracy.