Tactile sensing, encompassing the discernment and quantification of physical attributes including force, pressure, texture, shape, and temperature through direct contact with objects or surfaces, constitutes a fundamental cognitive ability inherent in both human and animal organisms. The human skin is equipped with various sensory receptors, including pain receptors, cold receptors, warm receptors, and mechanoreceptors that specialize in detecting different types of mechanical stimuli. Through these receptors, tactile sensing provides us with perceptible insights into the properties of materials and surfaces, particularly in terms of contact force and thermal characteristics. This wealth of tactile information not only aids our survival but also enables us to explore and respond to the external world in a more nuanced manner. This extraordinary capacity underpins a vast array of tasks, spanning from intricate manipulation and exploratory undertakings to sophisticated modes of communication. Moreover, tactile sensing extends profound relevance within the domain of artificial systems such as robots, prosthetics, and smart textiles [1,2,3]. By imbuing these systems with the capability to perceive and interpret tactile cues, their overall performance, functional versatility, and interactive potential can be significantly enhanced, propelling them to the forefront of technological innovation and human–machine interaction. Various types of tactile sensors have been developed to achieve better tactile sensing based on different materials and mechanisms. Among the various mechanisms employed, resistive, capacitive, piezoelectric, triboelectric, optical, and magnetic sensing principles have been extensively utilized. Each mechanism offers unique advantages and is suitable for specific applications.

Human skin is an exceptional organ comprising an integrated and stretchable network of sensors that relay information about thermal and tactile stimuli to the brain. This sensory network detects an array of characteristics, such as shapes, sizes, textures, and levels of contact pressure, allowing humans to maneuver safely and efficiently in their surroundings. Inspired by the sensory capabilities of human skin and leveraging advancements in artificial intelligence, materials science, and other technologies, researchers have made significant strides in the development of e-skin with human-like sensory abilities. E-skin can not only mimic the natural skin in certain aspects, but also integrate additional functionalities and required properties that surpass the natural skin. For instance, Rao et al. developed a tactile e-skin based on a single-electrode-mode TENG made of reduced graphene oxide and Bi₄Ti₃O₁₂ [108]. The e-skin not only possesses the ability to discriminate between temperature and pressure but also exhibits excellent flexibility. It can simultaneously detect and distinguish between temperature and pressure in real time with high sensitivity and a fast response time.

One of the major challenges in e-skin research is to achieve a self-healing function that can restore the mechanical and electrical properties of the damaged e-skin. Although some self-healing mechanisms have been developed, improvements are needed to ensure the long-term functionality and durability of e-skin under various environmental conditions. Duan et al. developed a water-modulated biomimetic hyper-attribute-gel e-skin (Hygel e-skin) that mimics the physical–chemical and sensory properties of human skin [109]. The Hygel e-skin possesses various skin-like attributes such as stretchability, self-healing, biocompatibility, biodegradability, weak acidity, antibacterial activities, flame retardance, and temperature adaptivity. It is capable of detecting and measuring pressure, temperature, humidity, strain, and contact with high resolution and accuracy. Moreover, it can also undergo reversible gel–solid transition via water or sweat modulation, enabling function reconfigurability and evolvability. The Hygel e-skin was attached to a robot to demonstrate its highly skin-like attributes in capturing multiple pieces of sensory information and reconfiguring required features, and its excellent skin compatibility for real-time gesture recognition through deep learning. Achieving a biomimetic e-skin that covers all the physicochemical and sensory properties of human skin, leading to more complex and versatile biomimetic applications, is the next-generation goal of artificial skin [110,111].

In recent years, human–machine interaction (HMI) has emerged as a promising field thanks to the rapid development of computer science, artificial intelligence, 5G communication, and biomimetic materials [112]. The traditional interaction paradigm based on the graphical interface has gradually evolved to incorporate multichannel input information and multimodal information expression. Various types of input information, such as vision, hearing, touch, language, expression, eye movement, gesture, and posture, are being utilized to interact with computer systems and enhance the user experience. Numerous studies have been conducted on the applications of HMI, with a particular focus on areas such as smart wireless control, intelligent keyboards, and robots with active perception.

Zhong et al. proposed a flexible tactile sensor array based on the textile woven structure by setting finger tapings and sounds to be signals as well as feedback, which can achieve a wearable HMI device similar to an electronic drum [40]. This device can realize single-point, multipoint, and sliding touch functions, which can be used for various HMI applications such as somatosensory games, intelligent robot systems, and artificial intelligence. Tao et al. developed a tactile hydrogel sensor with remarkable self-powered sensing capabilities, which functions as a switching button for the control of electrical appliances and robotic hands, mimicking human finger gestures [73]. Zhang et al. introduced a 3 × 3 cross-interactive tactile sensor array, which was successfully applied to effectively manipulate mechanical hand movements and keyboards in computer programs to play music [113]. The sensor array can achieve high output voltage and power density under low-pressure stimuli. Another example of a capacitive magnetic field/pressure sensor was proposed by Li et al., which can recognize various braille characters in a contactless manner through a flexible arrangement of permanent magnets [41]. Notably, the sensor was also proposed to realize visual sensing of the palm grasp state and strength feedback without an external power supply.

Tactile sensors have evolved beyond their traditional role of mimicking human skin and sensing external stimuli, which possess unique capabilities that surpass those of human skin, enabling them to monitor vital signs and contribute to personal health protection systems. By detecting and analyzing signals related to human blood pressure, heartbeat, and breathing rate, tactile sensors can identify and alert users or medical professionals about abnormal health conditions [114]. By utilizing these advanced capabilities, tactile sensors contribute to the development of personalized healthcare systems that empower individuals to monitor their health in real time. These sensors provide valuable data for the early detection and prevention of health issues, allowing for timely intervention and treatment. Furthermore, they can assist medical professionals in remote patient monitoring, providing patients’ vital signs. Chen et al. fabricated a capacitive tactile sensor that adopts a biomimetic hierarchical array architecture coated with silver nanowires [115]. This innovative sensor design enabled the real-time detection of large amounts of pressure, making it highly suitable for applications in healthcare monitoring and body motion recognition. The sensor can discern and differentiate subtle arterial pulse signals under various age, gender and motion conditions, as well as monitor the physiological activities under high pressure such as respiration and plantar pressure. Additionally, flexible bionic tactile sensors with high sensitivity and fast response based on the octopus suction cup microstructure were proposed [116]. The novel structure can endow the sensor with excellent sensing ability, making it an ideal capacitive sensor for the high-precision detection of objects grasped by bionic manipulators and wearable devices for monitoring human motion posture.