Dual-responsive fully self-healing triboelectric pressure sensor: Integrating truncated sphere morphology with deep learning-enhanced signal

We crafted a cutting-edge pressure sensor that exhibits exceptional dual-signal responsiveness, a testament to the innovative application of smart materials and artificial intelligence. This advancement is anchored by the MXene-dispersed hexafluoroisopropylene diphthalic anhydride-polydimethylsiloxane (M-FPP) composite, which is a truncated sphere-shaped material. This unique structure interacts with the graphene-coated polycaprolactone (G-PCL), adjusting its contact area to maintain accurate linearity across pressures ranging from 5 to 50 kPa. FPP and PCL, materials chosen for their self-healing properties, enable the sensor to recover its full structure and function after damage with targeted heat treatment. The sensor's advanced morphology elevates its ability to detect variations in both pressure and the duration of contact. Integration with a 1D-Convolution Neural Network (CNN) model enhances signal classification, achieving an impressive accuracy of 98.26 %. Such sensitivity allows for the recognition of minute differences in pressure and temporal parameters, surpassing conventional sensors' capabilities. With the M-FPP's design, the sensor's performance is significantly optimized, catering to varying pressure conditions. This blend of intelligent design and deep learning technology marks a stride toward adaptive and responsive sensing devices, capable of navigating the complexities of real-world stimuli with remarkable accuracy.