Data-driven Deep-sea Robot Technology Challenges and 
Development Path Exploration

Qilong Li

doi:10.70711/aitr.v2i10.7137

Data-driven Deep-sea Robot Technology Challenges and Development Path Exploration

Qilong Li

Abstract

Deep-sea robots are the core tools for exploring marine resources and carrying out scientific research, and the introduction of datadriven technology is promoting its transformation to the direction of intelligence and autonomy. This paper systematically analyzes the
application status of data-driven technology in the perception, decision-making and control of deep-sea robots, analyzes the key technical
challenges such as high-pressure environment data acquisition, small sample learning, and cross-modal fusion, and proposes a perceptionlearning-collaboration trinity development path. Combined with the frontier directions such as digital twin and edge computing, the future
breakthrough direction of deep-sea robot technology under multidisciplinary intersection is discussed.

Keywords

Data-driven; Deep-sea robot; Technical challenges; Development path

Full Text:

PDF

Included Database

References

[1] Van B, Lam T P, Huy N, et al. Data Augmentation Using Generative Adversarial Network for Automatic Machine Fault Detection Based

on Vibration Signals[J]. Applied Sciences, 2021, 11(5):2166-2166.

[2] Mao Z, Jagtap D A, Karniadakis E G. Physics-informed neural networks for high-speed flows[J]. Computer Methods in Applied Mechanics and Engineering, 2020, 360112789-112789.

[3] Erika C, Elisa D, Xiangpeng L, et al. Adaptive Extreme Edge Computing for Wearable Devices [J]. Frontiers in Neuroscience, 2021,

15611300-611300.

[4] Aoun G C, Mansour N, Dornaika F, et al. Environmental Constraints for Intelligent Internet of Deep-Sea/Underwater Things Relying on

Enterprise Architecture Approach.[J]. Sensors (Basel, Switzerland), 2024, 24(8):

[5] Y?ld?r?m F, Yalman Y, Bay?nd?r K, et al. Comprehensive Review of Edge Computing for Power Systems: State of the Art, Architecture, and Applications[J]. Applied Sciences, 2025, 15(8):4592-4592.

DOI: http://dx.doi.org/10.70711/aitr.v2i10.7137

Refbacks

There are currently no refbacks.