Design Principles for Building a Soft, Compliant, High Spatial Resolution Tactile Sensor Array

High-density tactile arrays are required to measure tactile properties, including forces and torques, contact shape and location, and dynamic slip, for dexterous gripping and manipulation tasks performed by robots and humans through haptic interfaces. How

  • PDF / 3,480,597 Bytes
  • 10 Pages / 439.37 x 666.142 pts Page_size
  • 68 Downloads / 233 Views

DOWNLOAD

REPORT


4

Graduate School of Biomedical Engineering, UNSW Australia, Kensington, NSW, Australia [email protected] 2 NeuRA, Randwick, NSW, Australia 3 Institute of Technical Physics, Riga Technical University, Riga LV1048, Latvia School of Medical Sciences, UNSW Australia, Kensington, NSW, Australia

Abstract. High-density tactile arrays are required to measure tactile properties, including forces and torques, contact shape and location, and dynamic slip, for dexterous gripping and manipulation tasks performed by robots and humans through haptic interfaces. However, in all current tactile sensing solutions, there is a trade-off between spatial resolution, flexibility, softness and manufacturing cost. In this work, a new design is proposed for a low cost, soft and malleable tactile sensing system with high spatial resolution that can be reshaped and applied to any surface, and that reduces the number of individual sensing elements, eliminates the need for any electronics within the sensing area and removes the need to time-division multiplex between sensor elements, allowing fully-parallel processing of transducer readings. Here the design is an orthogonal placement of conductive rubber strips with a pressure-dependant resistance. A basic algorithm for estimating the pressure at the intersection of each pair of orthogonal sensing strips is also described. The algorithm was tested with a simulated stimulation by two spherical stimuli onto a 16  16 grid (16 horizontal strips overlaid with 16 vertically strips) – the estimated pressure profile correlates well with the simulated stimulus (r = 0.86). A 5  5 grid prototype was built and was tested by stimulating with a spherical stimuli – the estimated pressure profile correlates well with the real stimulus (r = 0.92). Various design and algorithm improvements are suggested to overcome ambiguity in the estimated pressure profile due to the underdetermined nature of the system. Keywords: Force

 Tactile sensors  Sensor arrays  Robot sensing systems

1 Introduction Tactile sensing research has the potential to impact a large number of industries and disciplines [1], including robotics, telesurgery, intelligent hand/leg prostheses, computer peripherals, and haptic interfaces. High-density tactile arrays are required to © Springer International Publishing Switzerland 2016 F. Bello et al. (Eds.): EuroHaptics 2016, Part II, LNCS 9775, pp. 25–34, 2016. DOI: 10.1007/978-3-319-42324-1_3

26

H. Khamis et al.

measure tactile properties which cannot be determined by other sensing modalities such as vision (the current sensing method in robotic manipulators with dexterity comparable to humans for limited tasks) [2]. Robust and reliable tactile feedback of forces, torques, contact shape and location, and slip sensing are required for dexterous gripping and manipulation by robots and humans through haptic interfaces [1]. When replicating human dexterity, rigid sensor arrays are of limited practicality – unless the shape of a robotic/prosthetic finger matches the contact area, gripping a

Data Loading...