AiGIS and PyAiGIS: Tools for Geographic Visualization and Analysis for Irregular-shaped Small Bodies

We are developing a suite of tools for the visualization and analysis of exploration data from small celestial bodies with irregular shapes [1-8]. These tools represent the shapes of such bodies using polygon models and visualize them with 3D computer graphics. In addition, they can visualize various types of geospatial information associated with the shape models. We refer to this concept as 3D-GIS [7-8]. AiGIS and PyAiGIS are tools developed based on this concept. This paper reports the current development status of these two tools.

AiGIS is the first product in the software series based on the 3D-GIS concept. Written in Java, it is a standalone application that runs on major platforms including Windows, macOS, and Linux. It enables even beginners to easily visualize data and perform basic analyses (Fig. 1). AiGIS has been publicly available on our website for over five years[5-6] (AiGIS project site: https://arcspace.jp/aigis:top). While its core functionality is nearly complete, the software continues to receive updates to support new datasets and to fix bugs identified during use. The latest release was published in April, 2025.

Fig. 1. AiGIS screenshot. Shape and geographic information of the asteroid Ryugu are visualized.

PyAiGIS is a newly launched project aimed at developing an entirely new 3G-GIS environment based on Python and Jupyter Notebook [1-4]. Leveraging Python’s rich ecosystem of libraries, PyAiGIS is designed to enable interactive data manipulation, analysis, and visualization within the Jupyter Notebook environment (PyAiGIS project site: https://arcspace.jp/aigis2:top).

The system is built on PyVista, a powerful data visualization library [9]. PyVista serves as a Python interface for the Visualization Toolkit (VTK), an open-source software system widely used for manipulating and rendering polygon-based data [10]. PyVista provides access to VTK’s advanced visualization capabilities through a simplified and user-friendly interface, and allows for seamless handling of spatially referenced datasets. PyVista and VTK support the import of three-dimensional models in the Wavefront Object (OBJ) file format, which is the standard format used in AiGIS and similar tools for storing shape data of small bodies.

In AiGIS, geographic information is organized as a list of values corresponding to each plate of a three-dimensional polygonal shape model. While PyVista mesh objects can associate data directly with plates and vertices, PyAiGIS adopts a different design in which geographic data are managed separately using a Pandas DataFrame object. Pandas is a widely used Python library that provides robust data structures and flexible operations for handling numerical tables and time series [11]. Its capabilities for data selection, extraction, substitution, and statistical analysis make it particularly suitable for managing geospatial information.

To manage ancillary data and perform space geometry computations, PyAiGIS makes use of SpiceyPy [12], a Python wrapper for the SPICE toolkit developed by NASA’s Navigation and Ancillary Information Facility (NAIF). Although alternative libraries exist, SpiceyPy is chosen for its reliable and well-designed implementation, which ensures accurate and efficient handling of space geometry relevant to small body exploration data.

As outlined in our previous reports [1-4], the development of PyAiGIS follows a two-stage plan. The first stage involves experimental development through the creation of example Python scripts, while the second stage will focus on the practical implementation of reusable Python functions and data structures. The project is currently in the first stage, with ongoing efforts directed toward expanding the analysis and visualization capabilities of the PyAiGIS environment. The outcomes of this stage are being shared through the project’s GitHub repository (https://github.com/AiGIS-PyAiGIS/PyAiGIS).

The repository contains example code and documentation that demonstrate basic operations such as visualizing shape models and geographic data, extracting geospatial information stored in Pandas DataFrames, and generating and displaying latitude and longitude grids. It also provides example scripts for drawing mapping elements, such as lines and circles, at arbitrary locations on the surface of a shape model, as well as for overlaying global map images (Fig. 2). Furthermore, the repository offers guidance on selecting suitable visualization methods available in PyVista, depending on the user’s computing environment and intended use. Some procedures have already been implemented as Python functions, which are expected to serve as a foundation for the second stage of PyAiGIS development.

Fig. 2. PyAiGIS screenshots. Surface slope with color (A), geopotential height with contours (B), slope direction vectors (C), global mosaic image map (D), and line and circle drawing (E) are visualized on the asteroid Ryugu.

Acknowledgements: We thank T. Endo and T. Nagayoshi, who contributed to this work as an undergraduate/master’s student in our laboratory, for their efforts during the initial phase of the project. This project was supported by MEXT Promotion of Distinctive Joint Research Center Program and Promotion of Distinctive Joint Usage/Research Center Program (Grant Number: JPMXP0619217839, JPMXP0622717003 and JPMXP072383045) and supported by the JAXA Hayabusa2# International Visibility Enhancement Project.

References: [1] Hirata, N. et al. (2025) LPS LVI, Abstract #1873. [2] Hirata N. and Nagayoshi, T. (2024) LPS LV, Abstract #1877. [3] Nagayoshi, T. and Hirata, N. (2023) AGU Fall Meeting, P30F-3207. [4] Nagayoshi, T. and Hirata, N. (2022) AGU Fall Meeting, P25F-2187. [5] Hirata, N. et al. (2019) LPS L, Abstract #2347. [6] Hirata, N. et al. (2018) LPS XLIX, Abstract #1849. [7] Hirata, N. et al. (2008) LPS XXXIX, Abstract #1584. [8] Fujii, Y. et al. (2007) LPS XXXVIII, Abstract #1521. [9] PyVista, https://docs.pyvista.org/ [10] VTK, https://vtk.org/ [11] Pandas, https://pandas.pydata. org/ [12] Annex et al., (2020) Journal of Open Source Software, 5(46), 2050.