RNNs for Classification of Driving Behaviour

Abstract

Recurrent neural networks are an obvious choice for driving behavior analysis by means of time series of measurements, obtained either from telematics or mobile phone sensors. This work investigates such an application, employing two popular recurrent neural networks, i.e. long short-term memory networks and gated recurrent unit networks, as well as 1D convnets. Experiments are performed on a dataset comprising time series of measurements for four different types of driving. The results lead to the conclusion that gated recurrent unit networks achieve the highest classification accuracy, whereas they are more efficient than long short-term memory networks. Moreover, dropout and recurrent dropout lead to an approximately 3% increase with respect to classification accuracy. Naturally, 1D convnets are a more efficient neural network alternative at the cost of significantly lower classification accuracy.