Artificial intelligence: Digital smart pen helps people learn to write

Feb 09, 2023 Deixe um recado

Artificial intelligence: Digital smart pen helps people learn to write


A smart digital pen that helps people learn to write is currently being developed within the framework of the German-French project "Kaligo-based Intelligent Handwriting Teacher" (KIHT). The collaborative project, funded by the German Federal Ministry of Education and Research, is coordinated by the Karlsruhe Institute of Technology (KIT). In this project, KIT researchers studied artificial intelligence (AI) algorithms capable of reconstructing written marks and interpreting handwriting. Smart learning devices are here to benefit as many students as possible.



Handwriting is an important tool in the knowledge-based society. Repeated studies have shown that handwritten text is of higher quality than typed text. Now a new project aims to help people learn to write and combine handwriting with digital media. In the "Kaligo-based Intelligent Handwriting Teacher" (KIHT) project, research and industry partners from Germany and France have developed an intelligent digital pen. This innovative learning device is similar to a traditional pen in all major features and can be used to write on paper. However, it is equipped with an inertial sensor that captures the smallest changes in the free-space axis and thus can detect any position in space. The smart digital pen can be connected to any commercially available mobile end device, including tablets, and interact with the mobile Kaligo application. With the help of this app, exercises can be adjusted individually and data is automatically synchronized and stored.


smart writing pen


Teachers gain more freedom in creative and communicative tasks



"A smart digital pen, used together with an appropriate computer program, automatically supports learning to write. In this way, teachers and parents are freer for creative and communicative tasks," says Jürgen Becker, head of KIT's Institute for Information Processing Technologies (ITIV) said the professor. The software and AI algorithms are mainly handled by the French partners L'IRISA (Institut de Recherche en Informatique et Systèmes Aléatoires) and Learn&Go in Rennes. For example, Learn&Go developed the mobile Kaligo app for training writing. The integration of suitable AI concepts in embedded systems is the focus of the German partner KIT and the Heroldsberg-based writing instrument manufacturer STABILO International. The project is coordinated by KIT.



Researchers develop concept of ensemble algorithm



The group of Prof. Jürgen Becker at ITIV studies algorithms suitable for reconstructing written trajectories and interpreting handwriting. Researchers have developed various concepts to integrate AI algorithms as functions of embedded hardware. The complexity of the entire system is distributed between hardware and software, enabling fast and efficient execution of artificial intelligence. The aim is to reconstruct writing traces online from sensor data and develop efficient miniaturized hardware capable of interacting with all commercially available mobile end devices. At the end of the KIHT project, practical tests at demonstration plants are planned. Team leader Dr Tanja Harbaum (ITIV) highlighted in particular the great potential of the smart digital pen for schools: "This project brings together the capabilities of high-level international partners from science and industry and aims not only to advance research but also to provide Innovative products for children."