EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Motion Planning-Augmented Hierarchical Reinforcement Learning for Long-Horizon Mobile Manipulation

Hyungtai Kim iD, Mun-Taek Choi iD

Sensors · 17.06.2026 · 26 (12) · 3845

DOI: 10.3390/s26123845 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 17.06.2026. Verfasst von Hyungtai Kim und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute of Information & Communications Technology Planning & Evaluation, Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries, Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Long-horizon mobile manipulation requires a robot to execute a sequence of heterogeneous subtasks such as navigation, picking, and articulated-object manipulation in indoor environments. Standard reinforcement learning suffers from reward sparsity and inefficient exploration in this setting, and hierarchical methods often fail at the hand-off between consecutive subtasks when the terminal state of one subtask is kinematically infeasible for the next. We propose a motion planning-augmented hierarchical reinforcement learning architecture to resolve the fundamental trade-offs between sample efficiency and hand-off reliability in long-horizon mobile manipulation. The mission is decomposed into subtasks via a Semi-Markov Decision Process; within each subtask, a collision-free reference trajectory generated by RRT* in the full joint configuration space is embedded into the reward as a per-step shaping signal; and a region-goal mechanism, defined analytically from inverse kinematics feasibility, replaces rigid coordinate hand-offs with a continuous feasible region. The architecture is evaluated in the ManiSkill-HAB simulation under teleport-free sequential execution and challenging initialization. The proposed method improves subtask success rate and sample efficiency over the baseline across all six evaluated subtasks, and the advantage compounds along the long-horizon task chain.

Schlagwörter

Stichwörter der Autor:innen

mobile robotsmotion planningreinforcement learningrobot kinematicsservice robots

Autor:innen und Institutionen

  • School of Mechanical Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea.

ORCID-Profile: Hyungtai Kim, Mun-Taek Choi

Förderung

  • Institute of Information & Communications Technology Planning & Evaluation
  • Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries
  • Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry

Zitieren

APA

Kim, H., & Choi, M. T. (2026). Motion Planning-Augmented Hierarchical Reinforcement Learning for Long-Horizon Mobile Manipulation. Sensors, 26(12), 3845. https://doi.org/10.3390/s26123845

Vancouver

Kim H, Choi MT. Motion Planning-Augmented Hierarchical Reinforcement Learning for Long-Horizon Mobile Manipulation. Sensors. 2026;26(12):3845. doi:10.3390/s26123845

BibTeX
@article{Kim2026,
  author = {Kim, Hyungtai and Choi, Mun-Taek},
  title = {Motion Planning-Augmented Hierarchical Reinforcement Learning for Long-Horizon Mobile Manipulation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {12},
  pages = {3845},
  doi = {10.3390/s26123845}
}
Kostenlos starten →