Triboelectric Signal Dataset for Hybrid Deep Learning-Based Aerial Robot State Recognition

Description

This dataset provides triboelectric time-series sensing signals and deep-learning recognition results for an AI-assisted Meta zero-energy self-sensing aerial robot system. It includes horizontal vibration signals, tilt-motion signals, FEP-thickness-based sensor fault-diagnosis signals, processed classification samples, model-comparison results, and real UAV flight experimental data. The dataset supports the development and validation of a hybrid CNN-BiLSTM-GRU-attention model for robust aerial robot state recognition using self-powered triboelectric sensing signals under different laboratory and real-flight conditions.

Institutions

• Southwest Jiaotong University

  Sichuan, Chengdu

Categories

Licence