The desire to directly touch and experience virtual objects led to the development of a tactile feedback device. In this paper, a novel soft pneumatic actuator for providing tactile feedback is proposed and demonstrated. The suggested pneumatic actuator does not use an external air compressor but it is operated by internal air pressure generated by an electrostatic force. By using the actuator, we designed a glove to interact with virtual reality. The finger motions are detected by attached flexible piezoelectric sensors and transmitted to a virtual space through Bluetooth for interconnecting with a virtual hand. When the virtual finger touches the virtual object, the actuators are activated and give the tactile feedback to the real fingertip. The glove is made of silicone rubber material and integrated with the sensors and actuators such that users can wear them conveniently with light weight. This device was tested in a virtual chess board program, wherein the user picked up virtual chess pieces successfully.

Thermal perception is essential for the survival and daily activities of people. Thus, it is desirable to realize thermal feedback stimulation for improving the sense of realism in virtual reality (VR) for users. For thermal stimulus, conventional systems utilize liquid circulation with bulky external sources or thermoelectric devices (TEDs) on rigid structures. However, these systems are difficult to apply to compact wearable gear used for complex hand motions to interact with VR. Furthermore, generating a rapid temperature difference, especially cooling, in response to a thermal stimulus in real-time is challenging for the conventional systems. To overcome this challenge and enhance wearability, we developed an untethered real-time thermal display glove. this glove comprised piezoelectric sensors enabling hand motion sensing and flexible TEDs for bidirectional thermal stimulus on skin. The customized flexible TEDs can decrease the temperature by 10 °C at room temperature in less than 0.5 s. Moreover, they have sufficiently high durability to withstand over 5,000 bends and high flexibility under a bending radius of 20 mm. In a user test with 20 subjects, the correlation between thermal perception and the displayed object's color was verified, and a survey result showed that the thermal display glove provided realistic and immersive experiences to users when interacting with VR. People perceive various stimuli from the surrounding environment by using the sensory receptors of the body. This allows them to comprehend their situation and respond according to the surrounding environment, which facilitates survival in nature. With respect to virtual reality (VR) and augmented reality (AR), various types of wearable gear embedding sensors and actuators have been developed to provide stimulus from the virtual or augmented environment to users in the real world 1-3. To manipulate an object in VR and AR environments, a glove-type device is the most appropriate for the following reasons. A human hand has 27 degrees-of-freedom (DOFs), which is the highest number of DOFs among all body parts, thereby enabling the manipulation of objects with complicated forms. Moreover, according to the cortical homunculus of Penfield 4 , the hand, as an isolated part, is the most sensitive body part with different types of sensory receptors on the skin that enable us to feel multiple sensory modalities, including pressure, vibration, stretch, pain, and temperature. Furthermore, it is possible to perceive a wide range of stimuli ranging from light physical actions such as gentle touching to coarser actions such as pinching. To realize physical interaction with virtual objects, three major feedback displays have been employed in glove-type devices: tactile, haptic, and thermal 5. The tactile and haptic feedback displays employ micro motors and piezoelectric actuators, and they provide contact information by applying force and displacement to the mechanoreceptors in the skin of the hand. On the contrary, thermal feedback displays induce heat ...

Soft sensors are attracting significant attention in human–machine interaction due to their high flexibility and adaptability. However, estimating motion state from these sensors is difficult due to their nonlinearity and noise. In this paper, we propose a deep learning network for a smart glove system to predict the moving state of a piezoelectric soft sensor. We implemented the network using Long-Short Term Memory (LSTM) units and demonstrated its performance in a real-time system based on two experiments. The sensor’s moving state was estimated and the joint angles were calculated. Since we use moving state in the sensor offset calculation and the offset value is used to estimate the angle value, the accurate moving state estimation results in good performance for angle value estimation. The proposed network performed better than the conventional heuristic method in estimating the moving state. It was also confirmed that the calculated values successfully mimic the joint angles measured using a leap motion controller.