Research

Paper

TESTING February 23, 2026

Bellman Value Decomposition for Task Logic in Safe Optimal Control

Authors

William Sharpless, Oswin So, Dylan Hirsch, Sylvia Herbert, Chuchu Fan

Abstract

Real-world tasks involve nuanced combinations of goal and safety specifications. In high dimensions, the challenge is exacerbated: formal automata become cumbersome, and the combination of sparse rewards tends to require laborious tuning. In this work, we consider the innate structure of the Bellman Value as a means to naturally organize the problem for improved automatic performance. Namely, we prove the Bellman Value for a complex task defined in temporal logic can be decomposed into a graph of Bellman Values, connected by a set of well-known Bellman equations (BEs): the Reach-Avoid BE, the Avoid BE, and a novel type, the Reach-Avoid-Loop BE. To solve the Value and optimal policy, we propose VDPPO, which embeds the decomposed Value graph into a two-layer neural net, bootstrapping the implicit dependencies. We conduct a variety of simulated and hardware experiments to test our method on complex, high-dimensional tasks involving heterogeneous teams and nonlinear dynamics. Ultimately, we find this approach greatly improves performance over existing baselines, balancing safety and liveness automatically.

Metadata

arXiv ID: 2602.19532
Provider: ARXIV
Primary Category: cs.RO
Published: 2026-02-23
Fetched: 2026-02-24 04:38

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.19532v1</id>\n    <title>Bellman Value Decomposition for Task Logic in Safe Optimal Control</title>\n    <updated>2026-02-23T05:48:58Z</updated>\n    <link href='https://arxiv.org/abs/2602.19532v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.19532v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Real-world tasks involve nuanced combinations of goal and safety specifications. In high dimensions, the challenge is exacerbated: formal automata become cumbersome, and the combination of sparse rewards tends to require laborious tuning. In this work, we consider the innate structure of the Bellman Value as a means to naturally organize the problem for improved automatic performance. Namely, we prove the Bellman Value for a complex task defined in temporal logic can be decomposed into a graph of Bellman Values, connected by a set of well-known Bellman equations (BEs): the Reach-Avoid BE, the Avoid BE, and a novel type, the Reach-Avoid-Loop BE. To solve the Value and optimal policy, we propose VDPPO, which embeds the decomposed Value graph into a two-layer neural net, bootstrapping the implicit dependencies. We conduct a variety of simulated and hardware experiments to test our method on complex, high-dimensional tasks involving heterogeneous teams and nonlinear dynamics. Ultimately, we find this approach greatly improves performance over existing baselines, balancing safety and liveness automatically.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.RO'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.SY'/>\n    <published>2026-02-23T05:48:58Z</published>\n    <arxiv:primary_category term='cs.RO'/>\n    <author>\n      <name>William Sharpless</name>\n    </author>\n    <author>\n      <name>Oswin So</name>\n    </author>\n    <author>\n      <name>Dylan Hirsch</name>\n    </author>\n    <author>\n      <name>Sylvia Herbert</name>\n    </author>\n    <author>\n      <name>Chuchu Fan</name>\n    </author>\n  </entry>"
}