TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning

Gokul Puthumanaillam, Jae Hyuk Song, Nurzhan Yesmagambet, Shinkyu Park, Melkior Ornik
Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, PMLR 283:1484-1497, 2025.

Abstract

Autonomous agents operating in adversarial scenarios face a fundamental challenge: while they may know their adversaries’ high-level objectives, such as reaching specific destinations within time constraints, the exact policies these adversaries will employ remain unknown. Traditional approaches address this challenge by treating the adversary’s state as a partially observable element, leading to a formulation as a Partially Observable Markov Decision Process (POMDP). However, the induced belief-space dynamics in a POMDP require knowledge of the system’s transition dynamics, which, in this case, depend on the adversary’s unknown policy. Our key observation is that while an adversary’s exact policy is unknown, their behavior is necessarily constrained by their mission objectives and the physical environment, allowing us to characterize the space of possible behaviors without assuming specific policies. In this paper, we develop Task-Aware Behavior Fields (TAB-Fields), a representation that captures adversary state distributions over time by computing the most unbiased probability distribution consistent with known constraints. We construct TAB-Fields by solving a constrained optimization problem that minimizes additional assumptions about adversary behavior beyond mission and environmental requirements. We integrate TAB-Fields with standard planning algorithms by introducing TAB-conditioned POMCP, an adaptation of Partially Observable Monte Carlo Planning. Through experiments in simulation with underwater robots and hardware implementations with ground robots, we demonstrate that our approach achieves superior performance compared to baselines that either assume specific adversary policies or neglect mission constraints altogether. Evaluation videos and code are available at https://tab-fields.github.io.

Cite this Paper

BibTeX
@InProceedings{pmlr-v283-puthumanaillam25a, title = {TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning}, author = {Puthumanaillam, Gokul and Song, Jae Hyuk and Yesmagambet, Nurzhan and Park, Shinkyu and Ornik, Melkior}, booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference}, pages = {1484--1497}, year = {2025}, editor = {Ozay, Necmiye and Balzano, Laura and Panagou, Dimitra and Abate, Alessandro}, volume = {283}, series = {Proceedings of Machine Learning Research}, month = {04--06 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v283/main/assets/puthumanaillam25a/puthumanaillam25a.pdf}, url = {https://proceedings.mlr.press/v283/puthumanaillam25a.html}, abstract = {Autonomous agents operating in adversarial scenarios face a fundamental challenge: while they may know their adversaries’ high-level objectives, such as reaching specific destinations within time constraints, the exact policies these adversaries will employ remain unknown. Traditional approaches address this challenge by treating the adversary’s state as a partially observable element, leading to a formulation as a Partially Observable Markov Decision Process (POMDP). However, the induced belief-space dynamics in a POMDP require knowledge of the system’s transition dynamics, which, in this case, depend on the adversary’s unknown policy. Our key observation is that while an adversary’s exact policy is unknown, their behavior is necessarily constrained by their mission objectives and the physical environment, allowing us to characterize the space of possible behaviors without assuming specific policies. In this paper, we develop Task-Aware Behavior Fields (TAB-Fields), a representation that captures adversary state distributions over time by computing the most unbiased probability distribution consistent with known constraints. We construct TAB-Fields by solving a constrained optimization problem that minimizes additional assumptions about adversary behavior beyond mission and environmental requirements. We integrate TAB-Fields with standard planning algorithms by introducing TAB-conditioned POMCP, an adaptation of Partially Observable Monte Carlo Planning. Through experiments in simulation with underwater robots and hardware implementations with ground robots, we demonstrate that our approach achieves superior performance compared to baselines that either assume specific adversary policies or neglect mission constraints altogether. Evaluation videos and code are available at https://tab-fields.github.io.} }
Endnote
%0 Conference Paper %T TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning %A Gokul Puthumanaillam %A Jae Hyuk Song %A Nurzhan Yesmagambet %A Shinkyu Park %A Melkior Ornik %B Proceedings of the 7th Annual Learning for Dynamics \& Control Conference %C Proceedings of Machine Learning Research %D 2025 %E Necmiye Ozay %E Laura Balzano %E Dimitra Panagou %E Alessandro Abate %F pmlr-v283-puthumanaillam25a %I PMLR %P 1484--1497 %U https://proceedings.mlr.press/v283/puthumanaillam25a.html %V 283 %X Autonomous agents operating in adversarial scenarios face a fundamental challenge: while they may know their adversaries’ high-level objectives, such as reaching specific destinations within time constraints, the exact policies these adversaries will employ remain unknown. Traditional approaches address this challenge by treating the adversary’s state as a partially observable element, leading to a formulation as a Partially Observable Markov Decision Process (POMDP). However, the induced belief-space dynamics in a POMDP require knowledge of the system’s transition dynamics, which, in this case, depend on the adversary’s unknown policy. Our key observation is that while an adversary’s exact policy is unknown, their behavior is necessarily constrained by their mission objectives and the physical environment, allowing us to characterize the space of possible behaviors without assuming specific policies. In this paper, we develop Task-Aware Behavior Fields (TAB-Fields), a representation that captures adversary state distributions over time by computing the most unbiased probability distribution consistent with known constraints. We construct TAB-Fields by solving a constrained optimization problem that minimizes additional assumptions about adversary behavior beyond mission and environmental requirements. We integrate TAB-Fields with standard planning algorithms by introducing TAB-conditioned POMCP, an adaptation of Partially Observable Monte Carlo Planning. Through experiments in simulation with underwater robots and hardware implementations with ground robots, we demonstrate that our approach achieves superior performance compared to baselines that either assume specific adversary policies or neglect mission constraints altogether. Evaluation videos and code are available at https://tab-fields.github.io.
APA
Puthumanaillam, G., Song, J.H., Yesmagambet, N., Park, S. & Ornik, M.. (2025). TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning. Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, in Proceedings of Machine Learning Research 283:1484-1497 Available from https://proceedings.mlr.press/v283/puthumanaillam25a.html.

Related Material