An intrinsically embedded pressure-temperature dual-mode soft sensor towards soft robotics

Highlights

• The dual-mode soft sensor is intrinsically embedded in different soft robots.

• The dual-mode soft sensor can sense pressure and temperature simultaneously without mutual interference.

• The soft robots can recognize size, weight, temperature variation and hardness of the contact object.

• This work provides a new approach to integrate soft robots and soft sensors.

Abstract

Soft sensing is indispensable for precisely manipulate and recognize objects in soft robotics. However, it is a challenge to better integrate soft robots with soft sensors because the sensing performances of soft sensors will be induced by mechanical stress and deformation to change and degrade. Herein, we propose a new design of intrinsically embedded pressure-temperature dual-mode soft sensors in soft robots to sense external stimuli. The soft sensor consists of a skin-inspired structure, the chief material prepared is equivalent to that of the soft robot, with high sensitivity, excellent repeatability and robustness. It can perceive the pressure and temperature of objects simultaneously and independently without mutual interference. Furthermore, we demonstrated that the soft sensor integrates with different soft robots to recognize the size, weight, temperature variation and hardness of the contact object. This work provides a new approach to integrate soft robots and soft sensors, which has great potential in vast robot-human and robot-environment applications.

Introduction

Over the decades, soft robotics have attracted great research attention due to their remarkable achievements such as universal jamming gripper [1], multigait soft robot [2], worm robot [3], octopus robot [4], fully integrated soft octobot [5], growth robot [6], soft multilocomotion microrobot [7], and so on. Unlike their rigid counterparts, soft robots inspired by natural organisms consist of soft components [8], [9], [10], [11], [12], can be adaptively deformed and actuated in various extreme environments [10], [13], [14], and synergistically and safely collaborate with people, bridging the gap between human beings and machines [15], [16]. In particular, soft robotics require multiple perception capabilities that can opportunely perceive feedback and perform robot-environment interactions more safely and accurately. Although the actuation techniques of soft robotics have been extensively developed [17], [18], [19], the research on integration with soft physical sensors still lags behind. Compared with the skin of soft robotics, human skin can dynamically perceive interactive physical information in the surrounding environment. Inspired by the human skin, several sensing mechanisms were developed to transduce physical signals, such as pressure [20], temperature [21], strain [22] and shear force [23], into electrical [24] or optical signals [25]. Numerous soft physical sensors thus were reported and widely demonstrated in the technologies such as human-machine interaction [26] and artificial intelligence [27], and the applications such as health monitoring [27], [28] and prosthetic hands [29]. Among them, pressure and temperature sensors are the most basic ones for either human beings or soft robotics.

The past few years have witnessed the creation of the integration of soft physical sensors and soft robots based on different sensing mechanisms, simply by attaching the soft physical sensor to the surface of the soft robot, it can be used to sense information such as temperature and contact pressure [30], [31], [32], [33]. More notably, the sensor is embedded in the air cavity of the soft robot for sensing by 3D printing and other methods [34]. However, there are still several key limitations to the integration of soft sensors and soft robots that need to be resolved. So far, to enable pressure-temperature dual-mode sensing capability, although soft pressure and temperature sensors can be separately attached on the surface of soft robotic skin, which would overcomplicate the electrical wiring and system design [30], [31], [32], [33]. Although it is possible to transduce multiple stimuli into coupled signals in a single device so that it has the function of multi-functional sensing, it will be influenced by signals during the decoupling process. Moreover, special flexible multifunctional sensors also have the disadvantage that the output results are affected by other physiological information and high manufacturing costs [35], [36], [37]. Besides, the modulus incompatibility between soft sensors and soft robots will also damage sensing performances [29], [38].

In this article, taking advantage of the superior electrical and mechanical features of carbon material, we report a dual-mode soft sensor intrinsically embedded in different soft robots (soft robotic gripper, soft robotic hand, crawling soft robot). The dual-mode sensor consists of a sandwich structure composed of top and bottom electrodes and soft robot skin, which can sense the pressure through capacitive response, and the top electrode can sense temperature through resistance response. And the dual-mode sensor has a vertically stacked bimodal device configuration based on the separation between capacitance response and resistance response, avoiding decoupling the data, which can allow soft robots to detect pressure and temperature simultaneously and independently with only one sensor. Moreover, the intrinsically embedded manufacturing approach can seamlessly integrate the dual-mode soft sensor into soft robotic skin and effectively perceive interactive environment information. In addition, the soft sensor is made of flexible material (PDMS, Ecoflex), and its modulus is similar to that of the soft robot preparation material (Ecoflex), which ensures that the modulus is compatible.