Siamese network with dual attention for EEG-driven social learning: Bridging the human-robot gap in long-tail autonomous driving

• Published in: Expert systems with applications Vol. 291; p. 128470
• Main Authors: Zhou, Xiaoshan, Menassa, Carol C., Kamat, Vineet R.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2025
• Subjects: Attention mechanism; Autonomous vehicles; Cognitive robotics; Dynamic Time Warping (DTW); Electroencephalogram (EEG); Human-in-the-loop systems; Long-tail problem; Robotic driving; Siamese neural network; Long-tail problem; Attention mechanism; Human-in-the-loop systems; Siamese neural network; Dynamic Time Warping (DTW); Electroencephalogram (EEG); Cognitive robotics; Autonomous vehicles; Robotic driving
• ISSN: 0957-4174
• DOI: 10.1016/j.eswa.2025.128470

Robots with wheeled, quadrupedal, or humanoid forms are increasingly integrated into built environments. However, unlike human social learning, they lack a critical pathway for intrinsic cognitive development, namely, learning from human feedback during interaction. To understand human ubiquitous observation, supervision, and shared control in dynamic and uncertain environments, this study presents a brain-computer interface (BCI) framework that enables classification of Electroencephalogram (EEG) signals to detect cognitively demanding and safety–critical events. As a timely and motivating co-robotic engineering application, we simulate a human-in-the-loop scenario to flag risky events in semi-autonomous robotic driving—representative of long-tail cases that pose persistent bottlenecks to the safety performance of smart mobility systems and robotic vehicles. Drawing on recent advances in few-shot learning, we propose a dual-attention Siamese convolutional network paired with Dynamic Time Warping Barycenter Averaging approach to generate robust EEG-encoded signal representations. Inverse source localization reveals activation in Broadman areas 4 and 9, indicating perception–action coupling during task-relevant mental imagery. The model achieves 80% classification accuracy under data-scarce conditions and exhibits a nearly 100% increase in the utility of salient features compared to state-of-the-art methods, as indicated by integrated gradient attribution. Beyond performance, this study extends a theoretical discussion of how the proposed cognitive architecture for BCI agents can be extended, particularly incorporating the memory mechanism, to potentially help categorize diverse mental states and support both inter- and intra-subject adaptation. Overall, this research advances the development of cognitive robotics and socially guided learning for human-in-the-loop automation in complex built environments.