Software implementation of collecting and processing UAV telemetry data with streaming data transmission

This article discusses the development of a software system for simulating the generation of unmanned aerial vehicle telemetry and organizing its reliable transmission in real time with storage and visual interpretation of results. The proposed architecture includes a simulator written in C++, a message broker built on Apache Kafka, a message consumer written in Go, a PostgreSQL data storage system, and a stack for directly monitoring parameter changes using a combination of Prometheus and Grafana environments. The system was tested on generated data. It is shown that the developed solution ensures stable transmission and storage of telemetry data. Post-processing of the collected data and its display on a dashboard in the form of graphs and virtual instruments using Prometheus and Grafana were also implemented. The average message transit time from publication to writing to a table was less than a millisecond per point, which is within the acceptable limits for real-time telemetry processing systems. The obtained results can be used in the creation of telemetry collection systems for real UAVs in real time, including those operating in swarm mode.