论文标题
Emmi：善解人意的人机互动，用于建立自动驾驶的信任
EMMI: Empathic Human-Machine Interaction for Establishing Trust in Automated Driving
论文作者
论文摘要
储层计算是预测湍流的有力工具，其简单的架构具有处理大型系统的计算效率。然而，其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间，然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用，该系统模拟了湍流的若干特征。我们表明，使用径向基函数作为非线性投影器，即使只有部分观测并且不知道控制方程，也能稳健地捕捉复杂的系统非线性。最后，我们表明，当测量稀疏、不完整且带有噪声，甚至控制方程变得不准确时，我们的网络仍然可以产生相当准确的预测，从而为实际湍流系统的无模型预测铺平了道路。
Highly automated vehicles represent one of the most crucial development efforts in the automotive industry. In addition to the use of research vehicles, production vehicles for the general public are realistic in the near future. However, to fully exploit the benefits of these systems, it is fundamental that users have an appropriate level of trust in automation. Recent studies indicate that more research is needed in this area. Furthermore, beyond the management of user trust, the system should also convey a perceptible added value to realize not only trust, but also acceptance and thus use. The EMMI project pushes both, the management of user trust while conveying added value to the user. Therefore, an advanced socio-emotional user model for estimating user trust and various user-centered HMI systems with unique UX are being developed. Together, the systems are employed to induce changes in the users' trust in automated vehicles.
