The accurate and realistic transmission of tactile signals is essential for the advancement of haptic devices. Challenges arise due to the simultaneous presence of high-frequency tactile signals, such as surface textures, and low-frequency tactile signals, like shape or hardness, when interacting with real objects. This complexity is evident when trying to transmit various signals in existing studies such as single-actuator, multimodal, and hybrid haptic devices. This article introduces a novel hybrid haptic device that is soft and wearable, integrating pneumatic and electromagnetic actuators to improve the accuracy of tactile signal transmission. The pneumatic actuator handles low-frequency components, while the electromagnetic actuator is responsible for high-frequency components, each aligning with their specific mechanical properties. A series of experiments validate the mechanical properties of our proposed device, demonstrating its significant potential in enhancing the accuracy and realism of tactile signal transmission.
Proprioception is the sense of self-movement and body position. Because amputees doesn't have proprioception, hand amputees use their visual feedback to determine whether the robotic prosthetic hand actuated the desired finger movements. But the visual feedback differs when the vision of the user is poor or blocked, or when the user is in the dark. Therefore, for control of the robotic prosthetic hand, sensory feedback devices should be developed to deliver information on the behavior of the robotic prosthetic hand without visual feedback. To fulfill this needs, we propose a closed-loop integrated system consisting of the proprioceptive feedback device, electromyography (EMG) classification considering amputees, and robotic prosthetic control using convolution neural network(CNN) classification.