EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI-Driven Multimodal Fusion in Soft Robotics: From Heterogeneous Perception to Closed-Loop Control

Shui Guan

ITM Web of Conferences · 2026 · 89 · 02015

DOI: 10.1051/itmconf/20268902015 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ITM Web of Conferences, online erschienen am 14.08.2026. Verfasst von Shui Guan. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Soft robots have a very flexible structure, which poses severe challenges for perception systems. These challenges are the difficulties of sensor integration, nonlinear deformations and strong coupling between the multimodal signals. Therefore, in dynamic scenarios, single-modality perception is often not successful. Although, AI-based multimodal fusion has been revealed to be a helpful solution, current studies don't have a clear understanding of the evolution of these methods. The purpose of this paper is to review the multimodal fusion from a analytical framework. The analysis is carried out at three consecutive stages, 1) heterogeneous to homogeneous representation, 2) fixed to adaptive fusion, and 3) perception to closed-loop control. Within this framework, 6 kinds of methods are discussed, such as feature-level fusion, architecture level fusion, dynamic weight fusion, algorithm driven fusion, cross physical field fusion and application oriented fusion. According to the above feature alignment, adaptive fusion and perception-actuator coupling, the analysed research can be divided into the following three parts. Finally, the paper summarizes some of the key challenges and directions the work can take, such as unified representation, efficient models, and standardized evaluation.

Zitieren

APA

Guan, S. (2026). AI-Driven Multimodal Fusion in Soft Robotics: From Heterogeneous Perception to Closed-Loop Control. ITM Web of Conferences, 89, 02015. https://doi.org/10.1051/itmconf/20268902015

Vancouver

Guan S. AI-Driven Multimodal Fusion in Soft Robotics: From Heterogeneous Perception to Closed-Loop Control. ITM Web of Conferences. 2026;89:02015. doi:10.1051/itmconf/20268902015

BibTeX
@article{Guan2026,
  author = {Guan, Shui},
  title = {AI-Driven Multimodal Fusion in Soft Robotics: From Heterogeneous Perception to Closed-Loop Control},
  journal = {ITM Web of Conferences},
  year = {2026},
  volume = {89},
  pages = {02015},
  doi = {10.1051/itmconf/20268902015}
}
Kostenlos starten →