Processing and Simulation of Radiometric Observations Using the Open-Source TU Delft Astrodynamics Toolbox (Tudat)

Radiometric observations taken of a spacecraft allow for the precise reconstruction of its trajectory through the process of orbit determination. Besides the use in spacecraft navigation, this also allows for the estimation of related parameters such as ephemerides of natural objects, geophysical parameters, or fundamental physical quantities. Current state-of-the-art software for processing radiometric observations is typically closed source and not publicly available, which hinders the independent analysis, verification and interpretation of results. With new and upcoming missions providing more data in planetary systems that have been studied in previous missions, fusing information from multiple missions is becoming increasingly important, and will provide valuable insights in the evolution of planetary systems and constraints over longer timescales. To achieve this, a consistent and transparent combination of these different data sources is crucial to prevent unresolved biases or processing artifacts from affecting the solution. In this work we present the capabilities of the open-source TU Delft Astrodynamics Toolbox (Tudat) [1, 2] for processing and simulating radiometric observations from multiple missions and heterogeneous file formats, which forms the foundation for the integration of radiometric observations in the General Open Orbital Dynamics (GOOD) platform [3]. Applications of this framework to orbit estimation problems are currently underway (see [4, 5, 6]) and will be expanded in the future.

Various formats exist to store radiometric observations, including binary file formats, such as the ODF format (TRK 2-18), ATDF format (TRK 2-25), and TNF format (TRK 2-34), and plain text formats such as the IFMS format, and CCSDS TDM format. Currently, Tudat has readily available file readers for ODF files, TNF files (based on NASA Planetary Data System Software [7]) and IFMS files. Additionally, a generic text-file interface is available, allowing to ingest files based on a tabulated user-specified format, such as files containing processed radiometric range data used in planetary ephemerides [8]. To use the observations in orbit determination and parameter estimation, corresponding observation models are required that simulate the observable based on the current set of parameters. At present, Tudat has observation models implemented for two- and three-way sequential range and Doppler observations. Correction models for the observables can be specified by the user depending on the needs of their application. Available models include first- and second-order relativistic corrections [9], tropospheric corrections (ranging from simple models such as based on Estefan et al. [10] to high-fidelity approaches using the VMF3 model [11]), ionospheric corrections (ranging from tabulated models provided by the DSN [9] to high-fidelity models that use VTEC maps from IONEX files [12]) and charged particle corrections due to the Solar corona (using the model outlined in Verma et al. [13]).

Building upon previous work [2], we present the recently extended capabilities and future plans of Tudat in the processing and simulation of radiometric observations, with a focus on the unified processing, conversion to internal representations, and consistent application of user-defined observation corrections. Using pre-fit residuals (i.e., observations computed on the spacecraft reference trajectory, compared to the actual observed values) we validate recent developments to produce state-of-the-art results, and discuss future improvements. Previous and ongoing analyses with Tudat using data from NASA’s GRAIL, MRO and Cassini missions, and ESA Mars Express and Rosetta missions found the root mean square (RMS) of the two- and three-way Doppler pre-fit residuals in the order of several mHz at an integration time of 60 s. The sequential range pre-fits of MRO have an RMS of several meters, while retaining a small systematic trend [2]. Planned developments include the processing of ATDF files (using the tool by Verma [14]) and CCSDS TDM files, implementation of a one-way Doppler observation model, and implementation of tropospheric corrections based on Advanced Water Vapor Radiometer measurements. Combined with the capability of Tudat to process astrometric measurements in the same analysis (see [15, 16]), this lays the foundation for reproducible research in astrodynamics and planetary science. The Tudat development and core user team is committed to open science, by developing the tools fully open source, and disseminating full analysis scripts and data sets after publication. This will allow the community to reproduce, extend, and combine existing radiometric tracking-data analyses, to tackle challenges in data fusion and consistent data weighing of multi-mission, multi-datatype analyses in a transparent manner.

References
[1] Dirkx et al. EPSC. 2022.
[2] Gisolfi et al. IAC. 2025.
[3] Dirkx et al. EPSC. 2026.
[4] Hener et al. EPSC. 2026.
[5] Reichel et al. EPSC. 2026.
[6] Sánchez Rodríguez et al. EPSC. 2026.
[7] NASA Planetary Data System Software. PyTrk234. 2024.
[8] Verma. arXiv. 2014.
[9] Moyer. Wiley. 2003.
[10] Estefan et al. JPL Publication. 1994.
[11] Landskron et al. Journal of Geodesy. 2018.
[12] Hernández-Pajares et al. Journal of Geodesy. 2009.
[13] Verma et al. Astronomy & Astrophysics. 2013.
[14] Verma. SoftwareX. 2022.
[15] Revellino et al. EPSC. 2026.
[16] Attree et al. EPSC. 2026.