




Very Long Baseline Interferometry (VLBI) has revolutionized astrometry by accurately measuring the angular directions of compact extra-galactic radio sources, forming the celestial reference frame (CRF). The K-band (24 GHz) CRF, introduced in the multi-wavelength ICRF3, promises enhanced precision over lower frequencies but faces challenges like ionospheric effects and network geometry issues. The project KOSMIC, a consortium between ETH Zurich, HartRAO, and TU Wien, aims to address these challenges to enhance the precision and stability of the K-band CRF, impacting astronomy, geodesy, and space exploration. These efforts align with the IAU's ICRF maintenance call and support multi-wavelength CRF development.
In the first part of the project, we will develop a machine learning-based ionospheric model to optimally correct K-band VLBI observation for ionospheric effects. The model will rely on Global Navigation Satellite System (GNSS) data and be specifically tuned to perform well at the sites of the VLBI stations participating in K-band efforts. In the second part of the project, advanced scheduling techniques based on the VieSched++ software will be implemented in K-band observation plans to partially compensate for the limitations of the network geometry.
This contribution provides an overview of the project, outlining our strategies to address the objectives and highlighting the initial results achieved. Our new approaches regarding ionospheric corrections and scheduling will be operationally implemented to enhance future K-band observations and the corresponding CRF.