An important challenge that affects ultrasonic mid-air haptics, in contrast to physical touch, is that we lose certain exploratory procedures such as contour following. This makes the task of perceiving geometric properties and shape identification more difficult. Meanwhile, the growing interest in mid-air haptics and their application to various new areas requires an improved understanding of how we perceive specific haptic stimuli, such as icons and control dials in mid-air. We address this challenge by investigating static and dynamic methods of displaying 2D geometric shapes in mid-air. We display a circle, a square, and a triangle, in either a static or dynamic condition, using ultrasonic mid-air haptics. In the static condition, the shapes are presented as a full outline in mid-air, while in the dynamic condition, a tactile pointer is moved around the perimeter of the shapes. We measure participants’ accuracy and confidence of identifying shapes in two controlled experiments ($n_1 = 34, n_2 = 25$). Results reveal that in the dynamic condition people recognise shapes significantly more accurately, and with higher confidence. We also find that representing polygons as a set of individually drawn haptic strokes, with a short pause at the corners, drastically enhances shape recognition accuracy. Our research supports the design of mid-air haptic user interfaces in application scenarios such as in-car interactions or assistive technology in education.

中文翻译：

---

使用动态触觉指针对 2D 几何形状进行空中触觉渲染

与物理接触相比，影响超声波空中触觉的一个重要挑战是我们失去了某些探索性程序，例如轮廓跟踪。这使得感知几何特性和形状识别的任务更加困难。与此同时，人们对空中触觉及其在各种新领域的应用越来越感兴趣，需要更好地了解我们如何感知特定的触觉刺激，例如空中的图标和控制拨盘。我们通过研究在半空中显示 2D 几何形状的静态和动态方法来应对这一挑战。我们使用超声波空中触觉在静态或动态条件下显示圆形、正方形和三角形。在静态条件下，形状在半空中呈现为完整的轮廓，而在动态条件下，触觉指针围绕形状的周边移动。我们在两个对照实验（$n_1 = 34，n_2 = 25$）中测量参与者识别形状的准确性和信心。结果表明，在动态条件下，人们可以更准确地识别形状，并具有更高的信心。我们还发现，将多边形表示为一组单独绘制的触觉笔划，在角落处有短暂的停顿，可以大大提高形状识别的准确性。我们的研究支持在车载交互或教育辅助技术等应用场景中设计空中触觉用户界面。结果表明，在动态条件下，人们可以更准确地识别形状，并具有更高的信心。我们还发现，将多边形表示为一组单独绘制的触觉笔划，在角落处有短暂的停顿，可以大大提高形状识别的准确性。我们的研究支持在车载交互或教育辅助技术等应用场景中设计空中触觉用户界面。结果表明，在动态条件下，人们可以更准确地识别形状，并具有更高的信心。我们还发现，将多边形表示为一组单独绘制的触觉笔划，在角落处有短暂的停顿，可以大大提高形状识别的准确性。我们的研究支持在车载交互或教育辅助技术等应用场景中设计空中触觉用户界面。