EVOLVE:Control of evasive manoeuvres for automated driving: solving the edge cases, 项目负责人(PI),50万欧元,2021.01-2025.12
Project description: Automated Driving is a promising but also challenging area of innovation in the automotive industry. Despite the recent advances in deep learning for automated driving, hazardous driving scenarios such as evasive manoeuvres are “edge cases” where learning methods are less effective because representative data are statistically rare. In this project we will tackle these edge cases by means of an integrated physics plus data-driven learning approach, especially exploiting recent advances in tyre/road sensing technology to gain real-time information on vehicle states (e.g. tyre forces) and road conditions. The ultimate goal of the project is to develop novel adaptive and pro-active control and robust trajectory planning approaches that can deal with such variable and non-nominal vehicle/road conditions The position is motivated by a challenging practical problem, yet it requires a strong control-theory approach.