Textile Triboelectric Sensors Enabling Robotic Operations with Enhanced Actile and Proprioceptive Perception

The development of flexible triboelectric sensors for robotic hands is essential for enhancing tactile and proprioceptive perception in advanced robotic applications. Current robotic sensor systems often struggle with achieving human-like object recognition due to rigidity and limited sensing capabilities. This study optimizes sensor design by integrating PDMS as a tribonegative material and conductive fabric as a tribopositive material, creating triboelectric sensors capable of detecting both object shapes and surface materials. Machine learning (ML) will be employed to process multi-sensor data, enabling the robotic hand to recognize objects based on learned patterns from tactile and proprioceptive signals. The ML model will be trained using sensor data to enhance object recognition, grasping precision, and adaptive responses to varying surfaces and materials. This integration of ML will allow the robotic hand to achieve human-like dexterity, improving robotic manipulation and human-robot interaction. The flexible and scalable sensor design presents opportunities for broader implementation in automation and robotic systems requiring tactile feedback and subtle environmental interaction.