EUVIMEDEUVIMED entdecken
Crossref

Methods and algorithms for autonomous navigation of a mobile ground robot group

V.D. Solodov, M.M. Kolendovska, A.V. Kasparyants, P.V. Khramtsov, M.A. Pechenov, O.Yu. Sergiyenko

Radiotekhnika · 25.07.2026 · (225) · 127-146

DOI: 10.30837/rt.2026.2.225.13 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Radiotekhnika, online erschienen am 25.07.2026. Verfasst von V.D. Solodov und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The article presents a comprehensive analytical review of methods and algorithms for autonomous navigation of mobile ground robot groups. The architecture of a multi-level navigation system is considered, encompassing subsystems for sensory perception, localization and mapping, trajectory planning, and coordination of group robot behavior. Modern approaches to solving localization and environment mapping problems are analyzed, based on SLAM and collaborative C-SLAM algorithms that enable the joint construction of a global map by multiple robots. Considerable attention is given to sensor data fusion methods, in particular the Extended Kalman Filter and graph-based optimization methods. The main path planning algorithms are examined, including graph-based search methods A* and Dijkstra, sampling-based algorithms RRT and RRT*, the artificial potential fields method, the Dynamic Window Approach (DWA), and hybrid trajectory planning approaches. An analysis of the advantages and disadvantages of each approach is conducted from the perspective of route optimality, computational complexity, and suitability for use in multi-robot systems. Multi-agent path planning and inter-robot collision avoidance in dynamic environments are investigated separately. The paper also addresses modern methods for coordinating robot groups, including formation control algorithms, consensus protocols, and decentralized agent interaction mechanisms. The influence of communication network topology on group performance is analyzed, and conditions for ensuring system stability in the presence of communication delays and measurement noise are investigated. Significant attention is devoted to modern intelligent approaches based on machine learning and deep reinforcement learning, which enhance the adaptability of navigation systems in complex and uncertain environments. To evaluate algorithm performance, simulation experiments were conducted comparing A*, RRT*, and a hybrid A*+DWA approach in an environment with obstacles and multi-robot interaction. The results demonstrated that the hybrid approach provides the best trade-off between route length, motion smoothness, and navigation safety, while RRT* shows high planning speed and A* ensures stable optimal path finding on discretized maps. Promising directions for further research are identified, including the development of decentralized collaborative mapping systems, integration of multi-agent reinforcement learning, improved scalability of robot groups, and the application of semantic environment models for high-level planning.

Zitieren

APA

Solodov, V., Kolendovska, M., Kasparyants, A., Khramtsov, P., Pechenov, M., & Sergiyenko, O. (2026). Methods and algorithms for autonomous navigation of a mobile ground robot group. Radiotekhnika, 127-146. https://doi.org/10.30837/rt.2026.2.225.13

Vancouver

Solodov V, Kolendovska M, Kasparyants A, Khramtsov P, Pechenov M, Sergiyenko O. Methods and algorithms for autonomous navigation of a mobile ground robot group. Radiotekhnika. 2026:127-146. doi:10.30837/rt.2026.2.225.13

BibTeX
@article{Solodov2026,
  author = {Solodov, V.D. and Kolendovska, M.M. and Kasparyants, A.V. and Khramtsov, P.V. and Pechenov, M.A. and Sergiyenko, O.Yu.},
  title = {Methods and algorithms for autonomous navigation of a mobile ground robot group},
  journal = {Radiotekhnika},
  year = {2026},
  number = {225},
  pages = {127--146},
  doi = {10.30837/rt.2026.2.225.13}
}
Kostenlos starten →