<p dir="ltr">"Dataset AutoDyn: A Simulation-Based Temporal Shape-Based Learning Framework for Maneuver Recognition in Semi-Autonomous" Vehicles.</p>

The dataset was developed to support research in maneuver recognition, vehicle behavior modeling, and adaptive control systems for intelligent vehicles.

The data include synchronized measurements of speed (km/h), steering angle (°), lateral acceleration (m/s²), yaw rate (°/s), longitudinal acceleration (m/s²), throttle position (%), and brake pressure (bar) across multiple maneuver types such as lane change, cornering, merging, and obstacle avoidance. Each record represents a distinct maneuver instance sampled at consistent intervals during dynamic vehicle operation. This dataset is suitable for applications in: Temporal and spatial pattern recognition for vehicle maneuvers Training and benchmarking of machine learning models (e.g., LSTM, CNN, SVM) for trajectory classification Validation of driver-assistance and semi-autonomous vehicle control algorithms Development of predictive and adaptive control frameworks for vehicle dynamics All data are formatted in tabular form (Excel/CSV) and were generated from controlled simulation environments ensuring consistency and reproducibility.