Khaliliardali Zahra, Chavarriaga Ricardo, Andrei Gheorghe Lucian, Millán José del R
Defitech Chair in Non-Invasive Brain-Machine Interface, School of Engineering, École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, CH-1015.
Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:3829-32. doi: 10.1109/EMBC.2012.6346802.
Recognition of driver's intention from electroencephalogram (EEG) can be helpful in developing an in-car brain computer interface (BCI) systems for intelligent cars. This could be beneficial in enhancing the quality of interaction between the driver and the car to provide the response of the intelligent cars in line with driver's intention. We proposed investigating anticipation as the cognitive state leading to specific actions during car driving. An experimental protocol is designed for recording EEG from 6 subjects while driving the virtual reality driving simulator. The experimental protocol is a variant of the contingent negative variation (CNV) paradigm with Go and No-go conditions in driving framework. The results presented in this study support the presence of the slow cortical anticipatory potentials in EEG grand averages and also confirm the discriminability of these potentials in offline single trial classification with the average of 0.76 ± 0.12 in area under the curve (AUC).
从脑电图（EEG）识别驾驶员意图有助于开发用于智能汽车的车内脑机接口（BCI）系统。这对于提高驾驶员与汽车之间的交互质量、使智能汽车的响应符合驾驶员意图可能是有益的。我们建议将预期作为导致汽车驾驶过程中特定动作的认知状态进行研究。设计了一个实验方案，用于在6名受试者驾驶虚拟现实驾驶模拟器时记录脑电图。该实验方案是在驾驶框架中具有“执行”和“不执行”条件的关联负变化（CNV）范式的变体。本研究给出的结果支持脑电图总体平均值中存在缓慢的皮层预期电位，并且还证实了这些电位在离线单试验分类中的可区分性，曲线下面积（AUC）的平均值为0.76±0.12。
