Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications

2025-01-10

AI Summary

Scalable safety-critical coordinated control for large-scale multi-agent systems under Signal Temporal Logic (STL) specifications remains challenging due to prohibitive computational complexity in centralized optimization. Method: This paper proposes a decentralized graph-structured modeling framework that replaces conventional monolithic modeling and mixed-integer linear programming (MILP) planning with a graph neural network (GNN) to encode agent topology, integrated with distributed control barrier functions (CBFs) for collision avoidance and end-to-end joint optimization guided by STL constraints. Contribution/Results: Evaluated on systems with over one hundred agents, the method achieves more than a 5× speedup in planning time compared to state-of-the-art MILP approaches, while attaining STL satisfaction and collision-free navigation rates exceeding 98%. It effectively alleviates the computational bottleneck inherent in STL-based multi-agent synthesis, establishing a scalable paradigm for safety-guaranteed coordination in high-density robotic deployments.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Natural Language Processing: Safety and Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Existing methods for safe multi-agent control using logic specifications like Signal Temporal Logic (STL) often face scalability issues. This is because they rely either on single-agent perspectives or on Mixed Integer Linear Programming (MILP)-based planners, which are complex to optimize. These methods have proven to be computationally expensive and inefficient when dealing with a large number of agents. To address these limitations, we present a new scalable approach to multi-agent control in this setting. Our method treats the relationships between agents using a graph structure rather than in terms of a single-agent perspective. Moreover, it combines a multi-agent collision avoidance controller with a Graph Neural Network (GNN) based planner, models the system in a decentralized fashion, and trains on STL-based objectives to generate safe and efficient plans for multiple agents, thereby optimizing the satisfaction of complex temporal specifications while also facilitating multi-agent collision avoidance. Our experiments show that our approach significantly outperforms existing methods that use a state-of-the-art MILP-based planner in terms of scalability and performance. The project website is https://jeappen.com/mastl-gcbf-website/ and the code is at https://github.com/jeappen/mastl-gcbf .

Problem

• Multi-robot Coordination
• Signal Rules
• Computational Complexity

Innovation

• Graph Theory
• Signal Temporal Logic
• Autonomous Robot Navigation