Scalable Swarm Control Using Deep Reinforcement Learning
Bibliographic record
Abstract
Autonomous swarm navigation has been extensively studied due to its wide range of applications, from agriculture to surveillance and defense. Among the techniques used for swarm coordination, multi-agent reinforcement learning (MARL) has shown promise but is hindered by two main challenges. The first is the stochasticity of the environment, which is caused by the dynamic interaction among agents. The second is scalability, which becomes an issue as larger swarms require more complex neural networks and computational resources. To address these challenges, we propose a pipeline in which agents are trained using deep reinforcement learning in a single-agent, static environment. The resulting policy is then applied to multi-agent scenarios. We present a framework detailing this approach and its components. Our results show that a policy trained in a single-agent, static setting can be generalized effectively to multi-agent environments, mitigating the stochasticity issue. Furthermore, our model achieved collective behavior relying only on local information (agents in its neighborhood and a shared common goal), enabling scalability to large swarms. We compared our approach with a classical swarm control algorithm (flocking control) and a MARL approach (MADDPG), highlighting its efficiency and scalability. Our framework's performance is comparable with these two baselines, even performing better in some cases.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".