Despite its widespread use in signal collection, flexible sensors have been rarely used in human-machine interactions owing to its indistinguishable signal, poor reliability, and poor stability when inflicted with unavoidable scratches and/or mechanical cuts. A highly sensitive and self-healing sensor enabled by multiple hydrogen bonding network and nanostructured conductive network is demonstrated. The nanostructured supramolecular sensor displays extremely fast (ca. 15 s) and repeatable self-healing ability with high healing efficiency (93 % after the third healing process). It can precisely detect tiny human motions, demonstrating highly distinguishable and reliable signals even after cutting-healing and bending over 20 000 cycles. Furthermore, a human-machine interaction system is integrated to develop a facial expression control system and an electronic larynx, aiming to control the robot to assist the patient's daily life and help the mute to realize real-time speaking.

Progress toward the integration of electronic sensors with a signal processing system is important for artificial intelligent and smart robotics. It demands mechanically robust, highly sensitive, and self-healable sensing materials which could generate discernible electric variations responding to external stimuli. Here, inspired by the supramolecular interactions of amino acid residues in proteins, we report a self-healable nanostructured Ti3C2MXenes/rubber-based supramolecular elastomer (NMSE) for intelligent sensing. MXene nanoflakes modified with serine through an esterification reaction assemble with an elastomer matrix, constructing delicate dynamic supramolecular hydrogen bonding interfaces. NMSE features desirable recovered toughness (12.34 MJ/m3) and excellent self-healing performance (∼100%) at room temperature. The NMSE-based sensor with high gauge factor (107.43), low strain detection limit (0.1%), and fast responding time (50 ms) can precisely detect subtle human motions (including speech, facial expression, pulse, and heartbeat) and moisture variations even after cut/healing processes. Moreover, NMSE-based sensors integrated with a complete signal process system show great feasibility for speech-controlled motions, which demonstrates promising potential in future wearable electronics and soft intelligent robotics.

Supportinginformation and the ORCID identification number(s) for the author(s) of this article can be found under: https://doi.

We analyze the light scattered by a single InAs quantum dot interacting with a resonant continuous-wave laser. High resolution spectra reveal clear distinctions between coherent and incoherent scattering, with the laser intensity spanning over four orders of magnitude. We find that the fraction of coherently scattered photons can approach unity under sufficiently weak or detuned excitation, ruling out pure dephasing as a relevant decoherence mechanism. We show how spectral diffusion shapes spectra, correlation functions, and phase-coherence, concealing the ideal radiativelybroadened two-level system described by Mollow.