Snake-like robots face critical challenges in energy-efficient locomotion and smooth gait transitions, limiting their real-world deployment. This study introduces a bio-inspired compliant joint design integrated with a hierarchical neural oscillator network and an energy-optimized control framework. The joint mimics biological skeletal flexibility using specialized wheeled mechanisms and adaptive parallel linkages, while the control network enables adaptive gait generation and seamless transitions through a phase-smoothing algorithm. Critically, this work adopts a synergistic design philosophy where mechanical components and control parameters are co-optimized through shared dynamic modeling. The proposed predictive control strategy optimizes locomotion speed while minimizing energy consumption. Experimental simulations demonstrate that the method achieves an 18% higher average forward speed (0.0563 m/s vs. 0.0478 m/s) with 7% lower energy use (0.1952 J vs. 0.2107 J) compared to conventional approaches. Physical prototype testing confirms these improvements under real-world conditions, showing a 12.9% speed increase (0.0531 m/s vs. 0.0470 m/s) and 7.3% energy reduction (0.2147 J vs. 0.2317 J). By unifying mechanical flexibility and adaptive control parameter tuning, this work bridges dynamic performance and energy efficiency, offering a robust solution for unstructured environments.

Electroencephalogram (EEG)-based brain–computer interfaces (BCIs) offer a promising pathway for restoring communication. Decoding tonal languages like Mandarin from EEG remains challenging due to homophones and complex temporal dynamics. This study investigates the decoding of six high-frequency Mandarin action verbs—Chi (eat), He (drink), Chuan (wear), Na (take), Kan (look), and Dai (put on)—from EEG signals. We designed a visual-cue-based overt speech production experiment and collected EEG data from 30 participants during visually guided verb reading aloud. A recurrent neural network framework incorporating dual Long Short-Term Memory (LSTM) layers was implemented to model the long-range temporal dependencies in EEG patterns. The proposed model was compared against a traditional Common Spatial Pattern combined with Support Vector Machine (CSP-SVM) baseline. Our LSTM-based model achieved an average classification accuracy of 69.93% ± 3.07% for the six-class task, significantly outperforming the CSP-SVM baseline (36.53% ± 3.17%). Accuracy exceeded 75% under specific training conditions, including more than 15 training repetitions and a training-data proportion of 38%. Furthermore, the model attained this performance level utilizing approximately 38% of the available trial data for training, demonstrating data efficiency. The results indicate that the LSTM architecture can effectively capture the neural signatures associated with Mandarin verb processing, providing a foundation for developing practical EEG-based assistive communication technologies. The inference latency of the trained model, quantified as the post-training per-trial testing time, was under 2 s, supporting near-real-time applications.