PROJECT TITLE :

Powered Two-Wheeler Riding Pattern Recognition Using a Machine-Learning Framework

ABSTRACT:

During this paper, a machine-learning framework is employed for riding pattern recognition. The problem is formulated as a classification task to spot the class of riding patterns using data collected from three-D accelerometer/gyroscope sensors mounted on motorcycles. These measurements represent an experimental database used to investigate powered 2-wheeler rider behavior. Several well-known machine-learning techniques are investigated, including the Gaussian mixture models, the k-nearest neighbor model, the support vector machines, the random forests, and the hidden Markov models (HMMs), for both discrete and continuous cases. Additionally, an approach for sensor choice is proposed to spot the significant measurements for improved riding pattern recognition. The experimental study, performed on a true data set, shows the effectiveness of the proposed methodology and also the effectiveness of the HMM approach in riding pattern recognition. These results encourage the event of those methodologies within the context of naturalistic riding studies.