Driving Maneuvers Recognition and Classification Using A Hyprid Pattern Matching and Machine Learning
DOI: 10.14569/ijacsa.2023.0140230
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the critical need for automated driving behavior assessment to mitigate road accidents caused by human error and to support the development of advanced driver assistance systems. The authors propose a novel hybrid system for recognizing and classifying highway driving maneuvers using data collected from smartphone sensors. The study aims to overcome the limitations of existing methods, such as the computational burden of pure pattern matching techniques and the lack of interpretability in classical machine learning approaches. The experimental design utilized data from calibrated Android smartphones equipped with accelerometers, gyroscopes, and GPS, sampling at 50 Hz. Ten drivers performed six specific highway maneuvers—acceleration, braking, left/right lane changes, merging, and exiting—across light, normal, and hard intensity levels, generating 900 labeled samples. The system workflow involves signal filtering using a locally weighted running line smoother, coordinate reorientation, and maneuver detection via an adaptive sliding window algorithm based on short-term energy thresholds. The core comparison evaluates three classical machine learning classifiers (Random Forest, Support Vector Machine, and K-Nearest Neighbor) against a proposed hybrid approach. In the hybrid model, Dynamic Time Warping (DTW) is employed for maneuver recognition by comparing input signals to reference templates, while the machine learning algorithms handle the subsequent classification. The results demonstrate that the hybrid system outperforms the classical machine learning techniques. The study attributes this superior performance to the separation of the recognition and classification processes, which eliminates overlapping issues between target classes. By using DTW to measure signal similarity regardless of amplitude or duration variations, the hybrid approach provides more robust feature representation than the statistical features used in the standalone machine learning models. The evaluation metrics, including precision, recall, and F1-score, confirmed the enhanced accuracy of the hybrid method. The significance of this work lies in its contribution to intelligent transportation systems by providing a more accurate and computationally efficient method for driving behavior analysis. The findings suggest that integrating pattern matching with machine learning offers a viable solution for real-time monitoring of driving maneuvers using ubiquitous smartphone technology. This approach supports the broader goals of traffic safety, driver competency assessment, and the refinement of autonomous vehicle algorithms by providing clearer insights into driving patterns.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | canonical_url | — | — | 1 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | success | openalex | — | — | 1 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; 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