Robotic Sensing and Systems for Smart Cities

For several decades, various sensors and sensing systems have been developed for smart cities and civil infrastructure systems [...]

Bioinspired multisensory neural network with crossmodal integration and recognition

The integration and interaction of vision, touch, hearing, smell, and taste in the human multisensory neural network facilitate high-level cognitive functionalities, such as crossmodal integration, recognition, and imagination for accurate evaluation and comprehensive understanding of the multimodal world. Here, we report a bioinspired multisensory neural network that integrates artificial optic, afferent, auditory, and simulated olfactory and gustatory sensory nerves. With distributed multiple sensors and biomimetic hierarchical architectures, our system can not only sense, process, and memorize multimodal information, but also fuse multisensory data at hardware and software level. Using crossmodal learning, the system is capable of crossmodally recognizing and imagining multimodal information, such as visualizing alphabet letters upon handwritten input, recognizing multimodal visual/smell/taste information or imagining a never-seen picture when hearing its description. Our multisensory neural network provides a promising approach towards robotic sensing and perception.

Ear-Bot: Locust Ear-on-a-Chip Bio-Hybrid Platform

During hundreds of millions of years of evolution, insects have evolved some of the most efficient and robust sensing organs, often far more sensitive than their man-made equivalents. In this study, we demonstrate a hybrid bio-technological approach, integrating a locust tympanic ear with a robotic platform. Using an Ear-on-a-Chip method, we manage to create a long-lasting miniature sensory device that operates as part of a bio-hybrid robot. The neural signals recorded from the ear in response to sound pulses, are processed and used to control the robot’s motion. This work is a proof of concept, demonstrating the use of biological ears for robotic sensing and control.

Robotic Localization Based on Planar Cable Robot and Hall Sensor Array Applied to Magnetic Capsule Endoscope

Recently an active locomotive capsule endoscope (CE) for diagnosis and treatment in the digestive system has been widely studied. However, real-time localization to achieve precise feedback control and record suspicious positioning in the intestine is still challenging owing to the limitation of capsule size, relatively large diagnostic volume, and compatibility of other devices in clinical site. To address this issue, we present a novel robotic localization sensing methodology based on the kinematics of a planar cable driven parallel robot (CDPR) and measurements of the quasistatic magnetic field of a Hall effect sensor (HES) array. The arrangement of HES and the Levenberg-Marquardt (LM) algorithm are applied to estimate the position of the permanent magnet (PM) in the CE, and the planar CDPR is incorporated to follow the PM in the CE. By tracking control of the planar CDPR, the position of PM in any arbitrary position can be obtained through robot forward kinematics with respect to the global coordinates at the bedside. The experimental results show that the root mean square error (RMSE) for the estimated position value of PM was less than 1.13 mm in the X, Y, and Z directions and less than 1.14° in the θ and φ orientation, where the sensing space could be extended to ±70 mm for the given 34 × 34 mm2 HES array and the average moving distance in the Z-direction is 40 ± 2.42 mm. The proposed method of the robotic sensing with HES and CDPR may advance the sensing space expansion technology by utilizing the provided single sensor module of limited sensible volume.

Sensing skins for robots: product developments and recent research

Purpose This paper aims to provide details of the use of sensing skins by robots through reference to commercial products and recent research. Design/methodology/approach Following an introduction, this paper first summarises the commercial status of robotic sensing skins. It then provides examples of recent safety skin research and is followed by a discussion of processing schemes applied to multiple sensor skin systems including humanoid robots. Examples of research into soft, flexible skins follow and the paper concludes with a short discussion. Findings The commercialisation of sensing skins has been driven by safety applications in the emerging cobot sector, and a market is emerging for skins that can be retrofitted to conventional robots. Sensing skin research is widespread and covers a multitude of sensing principles, technologies, materials and signal processing schemes. This will yield skins which could impart advanced sensory capabilities to robots and potential future uses include agile manipulation, search and rescue, personal care and advanced robotic prosthetics. Originality/value This paper provides details of the current role of sensing skins in robots and an insight into recent research.