Machine Learning-based Trajectory Optimization of Connected and Autonomous Vehicles
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study addresses the challenge of accurately predicting vehicle trajectories in environments with Connected and Autonomous Vehicles (CAVs). While CAVs offer benefits such as improved safety, reduced emissions, and increased roadway capacity through coordinated movements, existing microscopic car-following models often fail to eliminate prediction errors regardless of parameter calibration. The research is motivated by the need to better understand CAV operations on freeways by leveraging machine learning, which can learn latent patterns from historical trajectory data more effectively than traditional parametric models. The methodology involves a comparative analysis between a machine learning approach and a traditional car-following model using historical trajectory data from the Next Generation SIMulation (NGSIM) database. Specifically, the study utilizes data from a freeway segment on US Highway 101 in Los Angeles. The researchers developed an XGBoost model, a supervised learning algorithm, to predict vehicle acceleration rates based on features such as reaction time, speed, headway, and acceleration. This was compared against the Intelligent Driver Model (IDM), a standard state-of-the-practice car-following model. The performance of both models was evaluated using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Additionally, the study conducted a comprehensive literature review of machine learning technologies, including supervised, unsupervised, and reinforcement learning, as well as traditional trajectory analysis methods. The results indicate that the XGBoost machine learning model outperformed the Intelligent Driver Model in predicting vehicle acceleration rates. The study identified specific feature importance rankings, determining which variables most significantly impact vehicle trajectory predictions. The comparison demonstrated that machine learning approaches could more accurately reproduce vehicle trajectories and predict upcoming states under varying traffic conditions compared to the traditional IDM. The findings provide a guideline for the prediction accuracy of vehicle trajectories using machine learning, highlighting the superiority of data-driven methods over conventional parametric models in this context. The significance of this research lies in its contribution to the development of more accurate trajectory prediction models for CAVs. By demonstrating the efficacy of machine learning in modeling longitudinal driving behaviors, the study supports the advancement of smart transportation systems. Accurate trajectory prediction is critical for mitigating traffic congestion, improving safety, and optimizing the performance of CAVs in freeway systems. The findings suggest that future research and implementation of CAV technologies should prioritize machine learning-based approaches to enhance the reliability and efficiency of autonomous vehicle operations.
Key finding
The XGBoost machine learning model outperforms the Intelligent Driver Model in predicting vehicle acceleration rates based on NGSIM trajectory data.
Methodology
dataset
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 24 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model