Teaching Kepler’s laws of planetary motion remains a persistent challenge in physics and astronomy education. These laws are often introduced through static diagrams and algebraic expressions that provide limited support for developing intuition about orbital motion, swept areas, and the relationship between orbital period and distance. As a result, many students struggle to connect mathematical formalisms with the physical behavior of planetary systems.

KeplerQuest-AI was created to address this challenge by integrating augmented reality (AR) and large language models (LLMs) within an inquiry-driven learning framework. The project enables learners to interact with three-dimensional representations of orbital mechanics while engaging in guided dialogue that supports reflection, hypothesis building, and conceptual understanding. Visualization and reasoning are treated as complementary processes, fostering deeper conceptual learning through exploration and inquiry.



The project is released as an open technological framework to promote transparency, reproducibility, and reuse in educational and research contexts. All software artifacts are hosted in the KeplerQuest-AI GitHub organization.

These artifacts are associated with a research study accepted for presentation at the 2026 IEEE Global Engineering Education Conference (EDUCON), where the pedagogical motivation, system design, and evaluation are discussed in detail.

From a technical perspective, KeplerQuest-AI is structured around two complementary and independent repositories. The augmented reality client is hosted in the repository keplerquest_ar, which contains the Unity-based AR application used to visualize and explore orbital mechanics related to Kepler’s laws. This repository focuses on the AR technology stack, including visualization components, interaction logic, and configuration for deployment.

The LLM-based conversational backend is hosted in the repository keplerquest_llm. This repository provides the backend infrastructure for the conversational agents that guide learners through inquiry-based interactions. It includes prompt management, local LLM integration, and retrieval-augmented generation mechanisms designed to scaffold reasoning rather than deliver direct answers.

Together, these repositories constitute the technological core of KeplerQuest-AI. While they include detailed documentation on dependencies, setup, and reproduction, the pedagogical design, learning activities, and experimental protocols are documented separately in the associated research publication.

Unless otherwise specified, all materials in this organization are released under the Creative Commons Attribution–NonCommercial 4.0 International (CC BY-NC 4.0) license. If you use or adapt any component of KeplerQuest-AI, please cite the corresponding paper presented at the 2026 IEEE Global Engineering Education Conference (EDUCON).

By making both the AR and AI components openly available, KeplerQuest-AI aims to contribute to ongoing efforts in physics education to meaningfully integrate immersive and intelligent technologies, supporting deeper understanding through interaction, visualization, and inquiry.

For more details, please refers to the paper (soon).