Electronic skin helps to adapt and interact with the environment through tactile stimulus. However, a conventional computing system can be inappropriate for processing signals generated from electronic skin, because the signals are enormous and change from time to time. Thus, attempts to overcome the problem have been continuing through a neuromorphic approach.
Here, we report an intelligent haptic perception device (IHPD) that combines an organic electrochemical transistor (OECT)-based synaptic device and a pressure sensor. The pyramid-patterned ion gel, which performs both the dielectric layer of the synaptic transistor and the pressure sensor, simplifies the structure of the device. In addition, the pyramid structure makes it possible to stepwise dope a channel of the synaptic device by pressure. Finally, the IHPD is capable of processing and learning through a selective and reversible transition of short-term plasticity (STP) and long-term plasticity (LTP). This allows the IHPD itself to derive various information such as amplitude, frequency, duration, and the total amount of pressure, and to perform error-tolerant perception and learning through tactile stimuli. This new type of tactile sensor might be a new possibility to solve the existing problem of electronic skin.