Adaptive Software Architecture for Remote Control of Robotic Logistics Systems
Vìsnik Nacìonalʹnogo unìversitetu "Lʹvìvsʹka polìtehnìka". Serìâ Ìnformacìjnì sistemi ta merežì · 22.06.2026 · 19 · 316-331
DOI: 10.23939/sisn2026.19.316 ↗Kurz gefasst
Fachartikel in der Zeitschrift Vìsnik Nacìonalʹnogo unìversitetu "Lʹvìvsʹka polìtehnìka". Serìâ Ìnformacìjnì sistemi ta merežì, online erschienen am 17.05.2026. Verfasst von Oleksandr Romaniak und 1 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben.
Abstract (Originalsprache)
This paper addresses the challenge of ensuring reliable remote control of autonomous and semi-autonomous robotic logistics platforms over long distances-up to hundreds of kilometers-in the complex and dynamic environment of the modern battlefield, where communication links are frequently degraded. Particular attention is given to critical challenges such as high latency, intermittent connectivity, severely limited communication channel bandwidth, and deliberate interference from electronic warfare (EW) systems. Based on an analysis of existing communication protocols and architectural approaches, the authors propose a conceptual model of a multilayered adaptive software architecture of a robotic logistics system specifically designed to operate effectively under these adverse conditions. The proposed architecture integrates three core components: (1) an adaptive video compression module that performs semantic segmentation to identify and prioritize Regions of Interest (ROI) for high-fidelity transmission while heavily compressing or omitting background content; (2) a distributed command buffering and prioritization system augmented with timestamps to preserve command relevance and enable trajectory extrapolation during outages; and (3) a deterministic mechanism for seamless transitions between operational modes-ranging from full teleoperation to semi-autonomous execution and full autonomy-based on real-time communication quality metrics and mission context. The scientific novelty of the work lies in the formalization of the control mode switching process as a finite state machine, the state of which is determined by the threshold values of communication metrics and tactical situational awareness. To validate the concept, a functional prototype of the software system was implemented using an open-source software stack, employing the MQTT protocol over an emulated satellite channel with dynamically varying parameters. The computational experiment was conducted under multiple scenarios, including simulated EW jamming and fluctuating communication-channel bandwidth conditions. Key performance metrics were evaluated, including average control command delivery latency, communication channel bandwidth utilization efficiency, subjective quality of experience (QoE) for video observation based on the mean opinion score (MOS) scale, and the percentage of lost or delayed control commands. The results of the experiment demonstrate a 40 % improvement in overall system resilience and a 60–95 % reduction in required video bandwidth compared to a baseline non-adaptive full-frame encoding (constant bitrate and constant frame rate), while maintaining sufficient situational awareness for operator decision-making. Critically, the system retains task execution capability even during prolonged communication blackouts.
Zitieren
APA
Romaniak, O., & Levus, Y. (2026). Adaptive Software Architecture for Remote Control of Robotic Logistics Systems. Vìsnik Nacìonalʹnogo unìversitetu "Lʹvìvsʹka polìtehnìka". Serìâ Ìnformacìjnì sistemi ta merežì, 19, 316-331. https://doi.org/10.23939/sisn2026.19.316
Vancouver
Romaniak O, Levus Y. Adaptive Software Architecture for Remote Control of Robotic Logistics Systems. Vìsnik Nacìonalʹnogo unìversitetu "Lʹvìvsʹka polìtehnìka". Serìâ Ìnformacìjnì sistemi ta merežì. 2026;19:316-331. doi:10.23939/sisn2026.19.316
BibTeX
@article{Romaniak2026,
author = {Romaniak, Oleksandr and Levus, Yevheniya},
title = {Adaptive Software Architecture for Remote Control of Robotic Logistics Systems},
journal = {Vìsnik Nacìonalʹnogo unìversitetu "Lʹvìvsʹka polìtehnìka". Serìâ Ìnformacìjnì sistemi ta merežì},
year = {2026},
volume = {19},
pages = {316--331},
doi = {10.23939/sisn2026.19.316}
}
EUVIMED