location: Current position: Home >> Scientific Research >> Paper Publications

基于改进HMM的驾驶疲劳险态识别方法

Hits:

Indexed by:期刊论文

Date of Publication:2018-03-21

Journal:大连理工大学学报

Volume:58

Issue:2

Page Number:194-201

ISSN No.:1000-8608

Key Words:驾驶疲劳;隐马尔可夫模型;前向后向算法;粒子群优化算法

Abstract:驾驶疲劳的产生是渐进的动态生成过程,基于隐马尔可夫模型(hidden Markov model,HMM)的相关研究需首先确定模型训练初值,且训练过程易陷入局部最优.基于此,通过在 HMM训练过程中引入粒子群优化(particle swarm optimization,PSO)算法对训练过程存在的上述问题进行了改进,并结合驾驶疲劳状态典型数据集对所提出的改进方法和前向后向算法(forward-backward (BW)algorithm)进行了详细对比.实验及分析测试结果表明,所提出的改进方法在驾驶疲劳预测结果准确性和稳定性上都优于BW算法.

Pre One:A Decision-making Method for Connected Autonomous Driving Based on Reinforcement Learning

Next One:An Application of Particle Swarm Algorithms to optimize Hidden Markov Models for Driver Fatigue Identification