Driving-style-oriented human-like automated driving and its verification

面向驾驶风格的仿人自动驾驶及其验证

• 批准号: RGPIN-2018-05348
• 负责人: Cao, Dongpu
• 金额: $ 1.97万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=646805
• 关键词: Driving; style; oriented; human; like

Analyses of road accident data have shown that there are more than 1.2 million road fatalities every year worldwide. In most of the traffic accidents, the driver behaviour or driver error is the most critical causal factor. These are calling for an emerging development of vehicle automation which provides not only the improved driving safety, but also the increased efficiency and mobility for children, elderly, disabled who are not able to drive.*******The currently existing automated vehicles developed by several car manufacturers provide a very encouraging future for automated driving, however, fully automated driving is still infeasible at a large scale before overcoming some important technological challenges. Studies have also shown that many users experience motion sickness while riding in an automated vehicle, which would indicate a considerable negative impact on the public acceptance and quick penetration of automated vehicles. Furthermore, higher-level automated vehicles would be expected to provide users sufficient flexibilities for doing work-related tasks in the vehicle on the road. In order to enhance the user-friendliness as well as in-vehicle task efficiency for automated vehicles, this proposed research aims to develop a novel framework of human-like automated driving that integrates driving style recognition and autonomous driving algorithms. The research outcomes will help to improve the user friendliness, comfort and also achieve personalization of automated vehicles.*******The proposed research program will initiate a new paradigm for human-like automated driving with new scientific advances and technological innovations. The framework effectively bridges the fundamental understanding and characterization of human driver driving styles and autonomous driving trajectory planning and tracking control, for achieving the ‘human-like' attribute in automated driving. The new paradigm will help to enable the car companies to develop new-generation of automated driving vehicles to be personalized and more user-friendly. Furthermore, the research outcomes of this program will assist to boost the public acceptance, offer recommendations related to user experience in automated vehicles for policy makers, and accelerate the penetration of automated vehicles. The research program will also offer an excellent HQP training opportunity for the four graduate students directly involved as well as other student members in the laboratory. ******

对道路交通事故数据的分析表明，全世界每年有120多万人死于道路交通事故。在大多数交通事故中，驾驶员行为或驾驶员失误是最关键的原因。这些都要求车辆自动化的新兴发展，不仅提供了改进的驾驶安全性，而且提高了儿童，老人，残疾人无法驾驶的效率和机动性。*******目前几家汽车制造商开发的自动驾驶汽车为自动驾驶提供了一个非常令人鼓舞的未来，但是在克服一些重要的技术挑战之前，全自动驾驶仍然无法大规模实现。研究还表明，许多用户在乘坐自动驾驶汽车时经历了晕动病，这将对公众接受和快速渗透自动驾驶汽车产生相当大的负面影响。此外，更高级别的自动驾驶汽车有望为用户提供足够的灵活性，让他们在路上完成与工作相关的任务。为了提高自动驾驶车辆的用户友好性和车内任务效率，本研究旨在开发一种集成驾驶风格识别和自动驾驶算法的新型类人自动驾驶框架。研究成果将有助于提高自动驾驶汽车的用户友好性、舒适性和个性化。*******该研究计划将以新的科学进步和技术创新开创类人自动驾驶的新范式。该框架有效地将人类驾驶员驾驶风格的基本理解和表征与自动驾驶轨迹规划和跟踪控制相结合，实现了自动驾驶中的“类人”属性。这种新模式将有助于汽车公司开发更个性化、更人性化的新一代自动驾驶汽车。此外，该项目的研究成果将有助于提高公众的接受度，为政策制定者提供与自动驾驶汽车用户体验相关的建议，加速自动驾驶汽车的普及。该研究计划还将为四名直接参与的研究生以及实验室的其他学生成员提供极好的HQP培训机会。＊＊＊＊＊＊