Neural Network-Based Redundancy Resolution for Human-Inspired Motion Planning in Collaborative Robotics
Robotics · 13.11.2026 · 15 (11) · 243
DOI: 10.3390/robotics15110243 ↗Kurz gefasst
Fachartikel in der Zeitschrift Robotics, online erschienen am 13.11.2026. Verfasst von Federico Lozer und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This paper presents a neural network-based approach designed to allow redundant robotic manipulators to replicate the arm motion of a human subject. The proposed approach uses a neural network trained on a custom-built motion tracking dataset to properly select the optimal kinematic configuration of the robot that emulates human movements. Human-inspired motions can enhance trust in human–robot collaboration, as the manipulator movements become easier for the human co-worker to interpret. The proposed approach is validated using a simulated Franka Emika Panda robot with seven degrees of freedom based on experimentally recorded human motions. The results demonstrate the feasibility of the proposed approach and its effectiveness in reproducing human motion.
Zitieren
APA
Lozer, F., Scalera, L., Gasparetto, A., & Brandstötter, M. (2026). Neural Network-Based Redundancy Resolution for Human-Inspired Motion Planning in Collaborative Robotics. Robotics, 15(11), 243. https://doi.org/10.3390/robotics15110243
Vancouver
Lozer F, Scalera L, Gasparetto A, Brandstötter M. Neural Network-Based Redundancy Resolution for Human-Inspired Motion Planning in Collaborative Robotics. Robotics. 2026;15(11):243. doi:10.3390/robotics15110243
BibTeX
@article{Lozer2026,
author = {Lozer, Federico and Scalera, Lorenzo and Gasparetto, Alessandro and Brandstötter, Mathias},
title = {Neural Network-Based Redundancy Resolution for Human-Inspired Motion Planning in Collaborative Robotics},
journal = {Robotics},
year = {2026},
volume = {15},
number = {11},
pages = {243},
doi = {10.3390/robotics15110243}
}
EUVIMED