Summary of Predicting Overtakes in Trucks Using Can Data, by Talha Hanif Butt and Prayag Tiwari and Fernando Alonso-fernandez
Predicting Overtakes in Trucks Using CAN Data
by Talha Hanif Butt, Prayag Tiwari, Fernando Alonso-Fernandez
First submitted to arxiv on: 8 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A machine learning model is designed to predict truck overtakes from CAN data, which is crucial for safe driving decisions. The approach uses three classifiers – Artificial Neural Networks (ANN), Random Forest, and Support Vector Machines (SVM) – to detect overtakes up to 10 seconds before the event. The analysis shows that prediction scores increase as the overtake trigger approaches, while the no-overtake class remains stable or oscillates depending on the classifier. The best accuracy is achieved when approaching the trigger, but early overtaking prediction remains challenging. The classifiers achieve high recall (over 93%) in classifying overtakes, but suboptimal accuracy (typically 80-90% and below 60% for one SVM variant) in classifying no-overtakes. Combining two classifiers improves no-overtake classification (TNR > 92%) at the expense of reducing overtake accuracy (TPR), providing a balanced performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Truck overtakes are important for safe driving, and predicting them early can help prevent accidents. Researchers used special data from trucks to train three different models to detect when a truck is going to pass another one. The models look at the truck’s systems up to 10 seconds before the overtake happens. They found that the models get better at guessing when an overtake will happen as it approaches, but they’re not perfect. The best model can correctly predict most overtakes (over 93%), but it’s not great at saying when no overtakes will happen (usually around 80-90%). By combining two of the models, researchers got even better at predicting no overtakes, but it didn’t help as much with predicting overtakes. |
Keywords
* Artificial intelligence * Classification * Machine learning * Random forest * Recall
