Recognition of driver emergency braking behaviour based on support vector machine optimised by memetic algorithm

Abstract

Surface electromyography (sEMG) is one of the main information sources of human motion detection and has been widely used. The lower limb sEMG signal is introduced into the recognition model of emergency braking behaviour, and the features from time domain, frequency domain and model parameters are extracted to construct a feature vector. The emergency driving behaviours are identified by using support vector machine (SVM), and a memetic algorithm (MA) based on particle swarm optimisation and hill climbing algorithm is proposed to optimise the parameters of SVM. The results show that the model has better classification performance than that without optimisation. The final recognition rate of emergency braking behaviour of same individual is up to 92.3%, and that of different individuals can reach 85.6%. Moreover, the system can detect emergency braking 220 ms earlier than operating brake pedal. At 100 km/h driving speed, the braking distance is reduced by 6.1 m.