Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications

Joe Eappen, Zikang Xiong, Dipam Patel, Aniket Bera, Suresh Jagannathan
Proceedings of The 8th Conference on Robot Learning, PMLR 270:3516-3535, 2025.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-eappen25a, title = {Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications}, author = {Eappen, Joe and Xiong, Zikang and Patel, Dipam and Bera, Aniket and Jagannathan, Suresh}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {3516--3535}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/eappen25a/eappen25a.pdf}, url = {https://proceedings.mlr.press/v270/eappen25a.html}, 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.} }
Endnote
%0 Conference Paper %T Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications %A Joe Eappen %A Zikang Xiong %A Dipam Patel %A Aniket Bera %A Suresh Jagannathan %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-eappen25a %I PMLR %P 3516--3535 %U https://proceedings.mlr.press/v270/eappen25a.html %V 270 %X 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.
APA
Eappen, J., Xiong, Z., Patel, D., Bera, A. & Jagannathan, S.. (2025). Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:3516-3535 Available from https://proceedings.mlr.press/v270/eappen25a.html.

Related Material