ONHS Report: DARPA-PS-27-06

Unison

Solicitation ID: DARPA-PS-27-06

Agency: DARPA/STO

Type: Solicitation

Deadline: 2026-12-04T17:00:00 ET

Eligibility: Maybe (Dual Approval Required)

Funding: N/A

Doc Type: PS

Clearance: CUI

Program Manager: Unison@darpa.mil

Last Updated: 2026-11-03 17:04

Analyzed File: DARPA-PS-27-06.pdf

Eligibility Reasoning

FFRDCs (1) must clearly demonstrate that the proposed work is not otherwise available from the private sector; and (2) must provide a letter, on official letterhead from their sponsoring organization, that (a) cites the specific authority establishing the FFRDC eligibility to propose to government solicitations and compete with industry, and (b) certifies the FFRDC’s compliance with the associated FFRDC sponsor agreement’s terms and conditions. ... DARPA will assess and determine FFRDC... eligibility on a case-by-case basis during the Abstract evaluation phase.

Note: This is an AI-generated summary...

Summary

The Defense Advanced Research Projects Agency (DARPA) Unison program seeks to develop and evaluate collaborative autonomy approaches where teams of autonomous agents (tens to hundreds) jointly adapt their decision-making (actions) and communication strategies during mission execution. The core hypothesis is that unifying action and communication planning in real-time will improve mission effectiveness, scalability, and resilience in dynamic, uncertain environments with constrained or degraded communications. The program is divided into two interdependent Technical Areas (TAs). TA-1, Joint Co-Adaptation, focuses on creating the autonomy software prototypes that dynamically balance computational tractability, communications efficiency, and operational resilience, with a key objective of reducing communication dependence by up to 50,000x compared to a baseline. TA-2, Modeling and Simulation Environment, will develop a common Faster-Than-Real-Time (FTRT, ≥100x) simulation testbed to train, integrate, and evaluate the TA-1 solutions across various mission scenarios. The program will use a modified version of the DARPA Contested Operations in Denied Environments (CODE) software as a government-furnished benchmark for performance comparison and testbed validation. (Source: DARPA-PS-27-06.pdf.)

Overall Technical Areas

Multi-Agent Collaborative AutonomyJoint Autonomy-Communications Co-adaptationDecentralized Decision Processes (Dec-POMDP)Faster-Than-Real-Time (FTRT) SimulationConstrained Communications ModelingMulti-Agent Reinforcement LearningSemantic and Pragmatic CommunicationsHierarchical and Federated LearningState Abstraction and RepresentationDynamic Coordination GraphsAdaptive Policy ExecutionSensor Data ModelingAdaptive Counter-AutonomyPNT-Denied Navigation

Probable LBNL Areas

BSACSAEESAETA

Focus Areas

Example Priorities:

Develop collaborative-autonomy planning and execution software prototypes that jointly adapt autonomy and communications during mission execution.
Enable autonomous teams to modify tactics, behaviors, coordination relationships, and information exchange as mission complexity and operational variability increase and communications become constrained and/or degraded.
Engineer, execute, and continuously adapt joint action–communications policies that dynamically balance computational tractability, communication efficiency, and operational resilience.
Investigate approaches like epistemic inference, pragmatic/semantic communications, factored Dec-POMDPs, coordination-graphs, state-abstraction, and hierarchical/federated/distributed learning.

Specific Technical Skills:

Multi-Agent SystemsDecentralized POMDPsReinforcement LearningSemantic CommunicationsDistributed OptimizationHierarchical PlanningFederated LearningState Abstraction MethodsCoordination AlgorithmsEpistemic Inference

Potential PIs for this Area:

Name	Score	Organization
Kameshwar Poolla	0.73	ETA \| Energy Technologies and Systems \| Energy Storage & Dist Dpt · Distributed Optimization
Alexandre Bayen	0.72	ETA \| Energy Analysis \| Systems & Energy Tech Analysis · State Abstraction Methods
Suren Byna	0.72	CSA \| SciData \| Data Science Research · Federated Learning
Prabal Dutta	0.72	ETA \| Building & Industrial Energy Systems \| Bldg Technologies Dept · Coordination Algorithms
Talita Perciano Costa Leite	0.71	CSA \| SciData \| Data & Computational Science · Reinforcement Learning
Claire Tomlin	0.71	BSA \| Biological Systems & Engineering \| Department of BioEngineering and BioMedical Sciences · Decentralized POMDPs
Houjun Tang	0.70	CSA \| SciData \| Data Science Applications · Multi-Agent Systems
Zixi Hu	0.70	CSA \| AMCR \| Mathematics · Epistemic Inference
Cristian Poliziani	0.70	ETA \| Energy Analysis \| Systems & Energy Tech Analysis · Hierarchical Planning
Arpit Gupta	0.70	CSA \| ESNET \| Tech Adv and Engagement · Semantic Communications

Example Priorities:

Develop a Faster-Than-Real-Time (FTRT) modeling and simulation prototype environment to train, integrate, evaluate, and stress-test autonomy approaches.
Achieve an execution factor of at least 100x real-time to enable repeated mission runs, training cycles, parameter sweeps, and stress testing.
Model Denied, Degraded, Intermittent, and Limited (DDIL) communications conditions, including bandwidth limits, latency, packet loss, and topology changes.
Apply fidelity selectively to represent factors that materially affect mission outcomes while allowing abstraction to reduce runtime.
Provide automation to script, configure, execute, and control scenarios, including changes in mission conditions and operational stressors.

Specific Technical Skills:

Modeling and Simulation (M&S)High-Performance Computing (HPC)Network SimulationCommunications ModelingAPI Design & IntegrationModular Software ArchitectureTest Automation & OrchestrationSensor ModelingEnvironment ModelingData Capture and Analysis

Potential PIs for this Area:

Name	Score	Organization
Erich Strohmaier	0.77	CSA \| AMCR \| Computer Science · High-Performance Computing (HPC)
Kaiyu Sun	0.76	ETA \| Building & Industrial Energy Systems \| Bldg Technologies Dept · Environment Modeling
Michael Sohn	0.75	ETA \| Energy Analysis \| Systems & Energy Tech Analysis · Sensor Modeling
Zachary Needell	0.74	ETA \| Energy Analysis \| Systems & Energy Tech Analysis · Modeling and Simulation (M&S)
Gregory Lemieux	0.74	EESA \| CESD \| Earth System Sciences · Modular Software Architecture
Koushik Sen	0.73	CSA \| AMCR \| Computer Science · Test Automation & Orchestration
Evangelos Chaniotakis	0.73	CSA \| ESNET \| Systems and Software · Network Simulation
Devarshi Ghoshal	0.72	CSA \| SciData \| Data Science Applications · Data Capture and Analysis
Seth Carbon	0.72	BSA \| Env. Genomics & Sys. Biology \| BioSystems Data Science · API Design & Integration
Zhi Jackie Yao	0.72	CSA \| AMCR \| Applied Mathematics · Communications Modeling

Potential LBNL PIs (Overall)

Note on PI Matching: These suggestions are generated through an AI-driven semantic analysis of LBNL staff profiles.

Name	Score	Organization
Dipak Ghosal	0.74	CSA \| ESNET \| Tech Adv and Engagement · Joint Autonomy-Communications Co-adaptation
Wei Zhang	0.74	CSA \| SciData \| Data Science Research · Sensor Data Modeling
Christian Schmid	0.73	ETA \| Building & Industrial Energy Systems \| Bldg Technologies Dept · Semantic and Pragmatic Communications
Alexandre Bayen	0.72	ETA \| Energy Analysis \| Systems & Energy Tech Analysis · Dynamic Coordination Graphs
Angelos Ioannou	0.72	CSA \| AMCR \| Computer Science · Faster-Than-Real-Time (FTRT) Simulation
Myles Collins	0.72	ETA \| Energy Analysis \| Energy Markets & Planning · Adaptive Policy Execution
Daniel Arnold	0.72	ETA \| Energy Technologies and Systems \| Energy Storage & Dist Dpt · Adaptive Counter-Autonomy
Satyarth Praveen	0.72	CSA \| SciData \| Data Science Applications · Hierarchical and Federated Learning
Claire Tomlin	0.71	BSA \| Biological Systems & Engineering \| Department of BioEngineering and BioMedical Sciences · PNT-Denied Navigation
Bruce Mah	0.71	CSA \| ESNET \| Tech Adv and Engagement · Constrained Communications Modeling

Fundamental Research Exemption (FRE)

All DARPA efforts, including those involving Fundamental Research, at a minimum require prime performers and subcontractors to demonstrate compliance with CMMC Level 1, which focuses on the protection of FCI and consists of the security requirements that correspond to the 15 basic safeguarding requirements specified in 48 CFR 52.204-21.