Parameter-Efficient Federated Cooperative Learning for 3-D Object Detection in Autonomous Driving
Bibliographic record
Abstract
In the rapidly evolving field of autonomous driving, accurately detecting and understanding dynamic environments remains a challenge. Federated learning (FL) offers a promising approach by integrating decentralized models from multiple connected autonomous vehicles (CAVs) to enhance the performance of deep-learning (DL)-based object detection methods. However, traditional FL faces hurdles, such as extensive data synchronization requirements, limited data variance, and high communication costs. This article introduces a federated cooperative learning framework that addresses these challenges by combining data from both CAVs and roadside units. The framework combines local cooperative perception with global FL through a parameter-efficient FL adapter and a lazy communication strategy, improving DL-based object detection capabilities across diverse driving scenarios while significantly reducing bandwidth requirements. We also present a novel multiagent-multitown dataset Vehicle-to-Everything-Fed, specifically developed to validate the effectiveness of our approach under various conditions. Notably, our framework retains 97.51% of the detection accuracy achieved by full-model FL, while utilizing only 1.4% of the bandwidth typically required, demonstrating substantial improvements over conventional FL strategies. This study underscores the potential of our tailored approach to substantially enhance autonomous vehicle technologies with minimal resource utilization.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".