In the development of our proprietary software and software-hardware complexes, we primarily utilize:
- Programming Languages: C++, JavaScript
- Development Methodologies: Agile, Scrum
- Operating Systems: Windows, Linux
- Development Environments: MS Visual Studio, NetBeans, CLion
- Graphics Libraries: MFC, Qt
- Distributed File Systems: CephFS, BeeGFS, GPFS
- Parallelization and Distributed Computing: OpenMP, OpenMPI, SLURM
- Database Management System: PostgreSQL
FUNCTIONAL COMPOSITION OF THE SOFTWARE PLATFORM:
Cluster Computing — automatic high-performance distributed processing of large data streams in a microservices architecture paradigm with maximum efficient utilization of available computing resources of the cluster or cloud data center.
Primary Processing — automatic unpacking of the raw dump session from the spacecraft (RAW), extracting the shooting routes, and generating Level 0 (L0) archived data based on these routes.
Data Quality Improvement — automatic end-to-end quality control at all stages of processing and adaptive data quality enhancements, for example, automatic refinement of the geodetic referencing of satellite imagery data to the meter level (1-2 GSD) by correcting the onboard ballistic and orientation parameters of the spacecraft at the moment of imaging.
Standard Processing — automatic processing of L0 archived routes and the production of standard Level 1 (L1, georeferenced) and Level 2 (L2, in cartographic projection) products based on them, delivered to consumers in the form of capture frames (scenes).
Derived Processing — automatic creation of derived Level 3 products, including high-precision seamless orthocorrected coverages (orthomosaics) over large areas.
Archiving and Cataloging — automatic maintenance of the archive and catalog of L0 routes, as well as the metadata storage of the remote sensing (geoportal) data publication subsystem .
KEY ADVANTAGES OF THE SOFTWARE PLATFORM:
Fully automated processing.
High processing accuracy due to the application of rigorous mathematical models and the elimination of root causes of errors rather than their consequences.
Consistently high quality of processing thanks to adaptive algorithms that mitigate onboard errors and end-to-end quality control at all stages.
High operational efficiency and performance of processing due to the use of distributed (cluster) computing technology. Maximum rational utilization of computing resources.
Maximum unification of all processes, formats, and metadata.
Compliance with international and industry standards.
Horizontal scalability to accommodate any large data streams from multi-satellite constellations (just by adding spacecraft).Possibility of functional expansion thanks to the platform approach and microservices architecture.
Portability across various computing platforms and data centers.
Minimal operational costs (only maintenance of spacecraft).