Investigation and Refinement of a Virtual Rover Navigation System in Unity Based on Reinforcement Learning
Received: 05 Oct 2026 Revised: 20 Oct 2026 Accepted: 08 Nov 2026
Published: 2026, vol. 30, issue 4, pp. 381–384
Abstract
We audit and refine a PPO-based virtual rover navigation prototype. The original observation lacked obstacle distances; the revised 52-dimensional vector contains 36 LiDAR features. Three independently trained policies were evaluated over 500 episodes each. The mean success rate was 46.8%, with a between-run standard deviation of 6.8 percentage points. The results apply to one simulated scene only.
Keywords: reinforcement learning, navigation, virtual rover, PPO, LiDAR.
BibTeX
@article{IS-Manosin-Kumskov2026,
author = {Manosin, Vladislav Valerievich and Kumskov, Mikhail Ivanovich},
title = {{Investigation and Refinement of a Virtual Rover Navigation System in Unity Based on Reinforcement Learning}},
journal = {Intelligent Systems. Theory and Applications},
year = {2026},
volume = {30},
number = {4},
pages = {381--384},
}
AMSBIB
\Bibitem{IS-Manosin-Kumskov2026}
\by V.\,V.~Manosin, M.\,I.~Kumskov
\paper Investigation and Refinement of a Virtual Rover Navigation System in Unity Based on Reinforcement Learning
\jour Intelligent Systems. Theory and Applications
\yr 2026
\vol 30
\issue 4
\pages 381--384
\lang In Russian
Published under
Creative Commons Attribution 4.0 International (CC BY 4.0)
RU