Tactile sensing in intelligent robotic manipulation – a review

Abstract

Purpose -When designing hardware and algorithms for robotic manipulation and grasping, sensory information is typically needed to control the grasping process. This paper presents an overview of the major grasping and manipulation approaches and the more common hardware used to obtain the necessary sensory information. Design/methodology/approach -This paper presents an overview of tactile sensing in intelligent robotic manipulation. The history, the common issues, and applications are reviewed. Sensor performance is briefly discussed and compared to the human tactile sense. Advantages and disadvantages of the most common sensor approaches are discussed. Some examples are given of sensors that are widely available as of today. Eventually, some examples of the state-of-the-art in tactile sensing application are presented. Findings -Although many sensor technologies and strong theoretical models have been developed, there is still much left to be done in intelligent grasping and manipulation. This is partly due to the youth of the field and the complex nature of safe control in uncertain environments. Even though there are impressive results when it comes to specific examples of advanced manipulation, there seems to be room for great improvements of hardware and especially algorithms when it comes to more generic everyday domestic tasks. Originality/value -This paper presents a review of sensor hardware while also giving a glimpse of the major topics in grasping and manipulation. While better hardware of course is desirable, the major challenges seem to lie in the development and application of grasping and manipulation algorithms.

Tactile sensors based on conductive polymers

“…They are demanded in applications where unstructured environments or uncertainty are present, like minimal invasive surgery (MIS), robotics, rehabilitation, virtual reality, telepresence, or industrial automation (Lee 2000;Tegin and Wikander 2005). Many different approaches have been proposed to fabricate these sensors, most of them are based on piezoresistive (Engel et al 2006; Kane et al 2000;Mei et al 2000;Lomas et al 2004;Kim et al 2006;Wisitsoraat et al 2007; Shan et al 2005) or capacitive (Salo et al 2003;Leineweber et al 2000;Paschen et al 1998;Gray and Fearing 1996;Lee et al 2006; http://pressureprofile.com/products-robotouch) principles, and a few are based on optical (Hellard and Russell 2006) or piezoelectrical transduction (Dahiya et al 2007).…”

Reliable non-prehensile door opening through the combination of vision, tactile and force feedback

“…It is known that there are about 17000 mechanoreceptors distributed along the fingers and the palm of the human hand, which provide rich information, mainly about the contact distribution, limb motion and forces (Howe, 1994). Several attempts to mimick the human sense of touch exist (Tegin and Wikander, 2005), being the tactile array sensors -able to perceive the pressure distribution of the contact and the local shape -the most common approach.…”

Object identification with tactile sensors using bag-of-features

“…Tactile sensors [12], [19] are commonly defined as "a device that can measure a given property of an object or contact event through physical contact between sensor and object" that is able to sense one or more of the following modalities: pressure, normal and sheer forces, torques, slip, vibrations, or temperature. Important properties of a sensor are its spatial and temporal resolution, noise, hysteresis, creep, and aging.…”