Recently, Professor Li Yang's team from the School of Integrated Circuits at Shandong University has made new advances in biomimetic electronic skin research. Their related findings, titled "Biomimetic Electronic Skin for Robots Aiming at Superior Dynamic-Static Perception and Material Cognition Based on Triboelectric-Piezoresistive Effects", have been published in the Nano Letters journal (SCI Q1, Impact Factor: 10.8). Professor Li Yang is the corresponding author of the paper, and doctoral student Li Hao is the co-first author. Shandong University is the first completion unit of the paper.

Flexible electronic skin (e-skin) that mimics human skin to perceive external stimuli has been considered a key technology to address the growing demand for the applications of robots. Although e-skin is quite a hot research topic, a fact is that most studies only focus on the detection of a single category of static or dynamic signal, which falls far short of what is required. Meanwhile, with the rapid development of e-skin and artificial intelligence, general tactile perception can no longer meet the application requirements of modern intelligent robots, and cognition capability is highly expected in the advancement of robots in the future. To date, the developed e-skins are significantly insufficient in terms of the system integration of robots and usability demonstration in practical robot applications, which is desperately awaited for the evolution of intelligent robots. Additionally, fabrication technologies such as lithography or nanoimprinting for preparing high-performance e-skins are complex and costly, and the e-skin for comprehensively mimicking human skin functions prepared by simple and cost-effective fabrication methods is necessary.

Given the above-mentioned challenges, this work demonstrates a type of novel biomimetic e-skin (denoted as BES) for robots to effectuate comprehensive tactile perception of both dynamic and static pressure through the conjunction of triboelectric and piezoresistive parts as well as an impressive cognition capability by further in combination with machine learning technology. In this work, (1) we propose to use the PVDF-HFP triboelectric layer and PVDF-HFP/PEDOT piezoresistive layer both with uniformity microstructures to realize the BES of high sensitivity as well as wide response range through a template method and a selected extra polymerization treatment, which is facile and cost-effective, thus overcoming the issues existing in the above-mentioned preparation processes. (2) To achieve simultaneous monitoring and decoding of dynamic and static signals, the independent triboelectric part and piezoresistive part are laminated and assembled, and as a result, the entire process of the robot grasping object was successfully observed by attaching the BES to the robot hand, which demonstrating excellent tactile perception ability comparable to human skin functions. (3) To demonstrate the practicability of the developed high-performance BES, which is also the most significant highlight of this work, a long short-term memory (LSTM) neural network model-based machine learning technique is introduced and applied to the BES to build up an intelligent material cognition system. The system is shown to be capable of realizing real-time recognition of 6 materials with indistinct morphology and smooth surface via one casual touch, with an accuracy of 96.8%, exhibiting superior cognition capability even than biological skin sensory system. It can be anticipated that the proposed flexible e-skin with specially introduced microstructures by the facile and low-cost preparation method, the dynamic and static signals simultaneous decoding method, and the intelligent material cognition system, bring new inspiration to realize smart robots with extraordinary perception ability as well as advanced cognition capability.