作者: Bao Rong Chang , Hsiu Fen Tsai , Chung-Ping Young
DOI: 10.1016/J.ESWA.2009.07.036
关键词:
摘要: In this study, fuzzy approach with fault-tolerance has proposed to fuse heterogeneous sensed data and overcome the problem of imprecise collision warning due perturbed input signal when processing pre-crash warning. Meanwhile, another relevant danger in drowsy driving, involving fatigue level, carbon monoxide concentration, breath alcohol was considered approximately reasoned an extra reaction time modify NHTSA algorithm. A vision-sensing analysis cooperating global-positioning system is applied for lane marking detection warning, particularly exchanging dynamic static information between neighboring cars via inter-vehicle wireless communications. addition event recording very useful accident reconstruction on scene also established here. order speed up fusion both quantum-tuned back-propagation neural network (QT-BPNN) adaptive network-based inference (ANFIS), a distributed dual-platform DaVinci+XScale_NAV270 been employed. Several tests system's reliability validity have done successfully, comparison effectiveness showed that our outperforms two current well-known collision-warning systems (AWS-Mobileye ACWS-Delphi).