Search Results

All drivers come with a driving signature during a driving. By aggregating adequate driving data of a driver via multiple driving sessions, which is already embedded with driving behaviors of a driver, driver identification task could be treated as a supervised machine learning classification problem. In this paper, we use a random forest classifier to implement the classification task. Therefore, we collected many time series signals from 60 driving sessions (4 sessions per driver and 15 drivers totally) via the Controller Area Network. To reduce the redundancy of information, we proposed a method for signal pre-selection. Besides, we proposed a strategy for parameters tuning, which includes signal refinement, interval feature extraction and selection, and the segmentation of a signal. We also explored the performance of different types of arrangement of features and samples.

This paper proposes an approach that uses the road preview data to optimize a shift schedule for a vehicle equipped with an automatic transmission. The road preview is inferred here from the so-called electronic horizon of a digital map that includes road attributes such as road grade, curvature, segment speed limit, functional class, etc. The optimized shift schedule selects the gear ratio whose optimization is conducted through applying a hybrid model predictive control method to the powertrain system, which is modelled as the multiple plants associated with multiple gears together with engine models. The goal of this optimization of shift schedule includes improving real world fuel economy and drivability. The real-world fuel economy gains using the proposed approach are achieved through optimizing gear ratio w.r.t. the road grade variations of the road ahead.