Robotics Research Hub
ORCH: Organizational Principles Enable Collective Intelligence in Embodied AI

ORCH: Organizational Principles Enable Collective Intelligence in Embodied AI

Zhengran Ji, Jonathan Hyun, Boyuan Chen · N/A · 2026

reinforcement-learning
PDF arXiv

Framework

N/A

License

N/A

Stars

N/A

Summary

Collective intelligence depends not only on the capabilities of individual members, but also on how those members are organized.

Abstract Summary

Collective intelligence depends not only on the capabilities of individual members, but also on how those members are organized. Yet artificial multi-agent systems are typically assembled using fixed organizational structures, even when the physical tasks they perform impose fundamentally different coordination requirements. Here we show that principles from human organization theory can be operationalized to organize large, heterogeneous collectives of embodied artificial agents. We introduce ORCH (Organizing Roles and Coordination Hierarchies), which constructs task-specific hierarchical organizations by combining pooled interdependence for work that can proceed concurrently with sequential interdependence for work governed by prerequisite relationships. Across 25 wildfire-response missions spanning reconnaissance, rescue, transportation, resource management, containment and suppression, we evaluated teams of up to 50 heterogeneous agents using eight large language models. Organizations constructed using these principles consistently outperformed four representative embodied multi-agent approaches across mission outcome, execution efficiency, exploration and computational resource use. Human-designed ORCH organizations improved final score by 63.97% and execution efficiency by 74.29% on average relative to the four prior frameworks. Organizations generated automatically by language models improved these measures by 43.63% and 52.53%, respectively. These advantages persisted across missions and underlying language models. Notably, collective performance was not monotonically determined by model scale. Analysis of long-horizon missions showed that hierarchical organization enabled teams to preserve concurrent activity within specialized groups while coordinating ordered transitions between mission phases.

Key Points

  • Collective intelligence depends not only on the capabilities of individual members, but also on how those members are organized.
  • Yet artificial multi-agent systems are typically assembled using fixed organizational structures, even when the physical tasks they perform impose fundamentally different coordination requirements.
  • Here we show that principles from human organization theory can be operationalized to organize large, heterogeneous collectives of embodied artificial agents.
  • We introduce ORCH (Organizing Roles and Coordination Hierarchies), which constructs task-specific hierarchical organizations by combining pooled interdependence for work that can proceed concurrently with sequential interdependence for work governed by prerequisite relationships.
  • Across 25 wildfire-response missions spanning reconnaissance, rescue, transportation, resource management, containment and suppression, we evaluated teams of up to 50 heterogeneous agents using eight large language models.

Results

We evaluated ORCH on all 25 CREW-Wildfire missions using five random seeds per task and eight LLMs as the base model for each agent including ChatGPT-5.4 [ 36 ] , Gemma-4-it [ 37 ] , Llama-4-Scout-16E-it [ 5 ] , ERNIE-4.5 [ 2 ] , Qwen-3.6 [ 47 ] , Nemotron-3-Super [ 3 ] , GLM-5.1 [ 15 ] , and DeepSeek-V4-Pro [ 8 ] . These models range in size and capabilities. They also include both proprietary and open-weight models.

Results — ORCH: Organizational Principles Enable Collective Intelligence in Embodied AI
Figure 2 : Normalized Aggregated Performance on 25 CREW-Wildfire Tasks, Ranking Analysis, and ANOVA Test Results . The detailed normalization process can be found in Section S3 – S4 in Supplementary Text. (A) ORCH with human-expert-designed and LLM-generated team hierarchies both significantly outperform the other four baselines in terms of output token usage, final score, and efficiency, and they rank top-3 in terms of exploration, input token usage, and the number of API calls. (B) ORCH has the highest proportion of first-place rankings across all the LLM-task-seed combinations for both final score and efficiency metric. (C) The ANOVA test results showed that the interaction term between algorithm and LLM is not statistically significant, while the remaining terms are.

Method

Previous embodied LLM-based multi-agent algorithms relied on fixed and naive team hierarchies, typically following an orchestral style in which a single manager supervises all worker agents [ 44 , 51 , 7 , 17 ] . In ORCH , we propose adaptive organizational hierarchies that are tailored to the task difficulty and the capabilities of the available worker agents.

Method — ORCH: Organizational Principles Enable Collective Intelligence in Embodied AI
Figure 3 : Normalized Aggregated Performance on 25 CREW-Wildfire Tasks by Language Models . The detailed normalization process can be found in Section S3 – S4 in the Supplementary Text. (A) Aggregated LLM performance on the 25 CREW-Wildfire tasks with ORCH with human-expert-designed team hierarchies. (B) Aggregated LLM performance on the 25 CREW-Wildfire tasks with ORCH with LLM-generated team hierarchies. Surprisingly, the middle-sized models perform better than large-sized models with ORCH under both organizational designs.

Sources and Demonstrations

Figures are reproduced from Ji et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.

Share

Related Papers

4D Radar Perception Algorithms for Autonomous Driving: A Review
arXiv preprint

4D Radar Perception Algorithms for Autonomous Driving: A Review

Xumin Wu, Jun Zhou, Jilin Mei et al.

Research on 4D millimeter-wave radar perception algorithms has flourished in recent years, extending from signal processing and object detection to semantic segmentation, motion estimation, occupancy prediction, and dynamic scene reconstruction.

vision reinforcement-learning benchmark
PDF Beginner