EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Autonomous-Exploration Method for Mobile Robots Based on Distributional Soft Actor–Critic

He Ding, Benke Gao, Hao Chen, Shuai Yang, Zan Tang, Hanqiang Deng, Quan Liu, Jian Huang

Applied Sciences · 11.06.2026 · 16 (12) · 5892

DOI: 10.3390/app16125892 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 11.06.2026. Verfasst von He Ding und 7 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Laboratory of Equipment State Sensing and Smart Support, Innovation Research Foundation of National University of Defense Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Autonomous exploration requires reliable viewpoint evaluation from partial observations. Existing reinforcement-learning methods commonly approximate each action by using a scalar expected return, which is inadequate in unknown environments with heterogeneous topology. Similar local observations may correspond to different hidden structures, causing the same action to produce divergent information gains and travel costs. This perceptual aliasing makes value estimation uncertain and can propagate biased decisions through temporal-difference learning. To address this problem, we propose a distributional-value learning framework for autonomous exploration. The core idea is to replace scalar action evaluation with return-distribution estimation, where the distribution mean measures long-term utility and the variance characterizes uncertainty induced by hidden topology. Based on this formulation, task-oriented value learning is developed to stabilize target estimation and suppress optimistic viewpoint evaluation under conditions involving uncertain returns. Simulation experiments indicate that the proposed adaptation can reduce redundant motion and improve exploration efficiency over representative baselines in the tested scenarios, while a preliminary real-robot deployment suggests the feasibility of transferring the learned policy to a physical platform.

Förderung

  • National Key Laboratory of Equipment State Sensing and Smart Support
  • Innovation Research Foundation of National University of Defense Technology

Zitieren

APA

Ding, H., Gao, B., Chen, H., Yang, S., Tang, Z., Deng, H., Liu, Q., & Huang, J. (2026). An Autonomous-Exploration Method for Mobile Robots Based on Distributional Soft Actor–Critic. Applied Sciences, 16(12), 5892. https://doi.org/10.3390/app16125892

Vancouver

Ding H, Gao B, Chen H, Yang S, Tang Z, Deng H, et al. An Autonomous-Exploration Method for Mobile Robots Based on Distributional Soft Actor–Critic. Applied Sciences. 2026;16(12):5892. doi:10.3390/app16125892

BibTeX
@article{Ding2026,
  author = {Ding, He and Gao, Benke and Chen, Hao and Yang, Shuai and Tang, Zan and Deng, Hanqiang and Liu, Quan and Huang, Jian},
  title = {An Autonomous-Exploration Method for Mobile Robots Based on Distributional Soft Actor–Critic},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {12},
  pages = {5892},
  doi = {10.3390/app16125892}
}
Kostenlos starten →