I-Corps: Cyber Tactile Perception Platform for Manufacturing Robotics Applications

Massachusetts Institute Of Technology, Cambridge MA

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the development of tactile sensing technologies for commercial applications in healthcare, workplace safety, and manufacturing. The proposed technology is designed to enable remote capture of information concerning activity and health risks of the wearer, e.g., remote screening of neuromuscular disorders leading to falls among older adults. Addressing this issue is vital as falls result in severe injuries, impacting independence and incurring substantial medical costs for insurers. In addition, the proposed technology may be used as a system facilitating remote monitoring and reporting of workplace injuries. Such a system may enhance workplace safety measures, reduce hazards, and improve insurance coverage pricing accuracy. There also is a potential need in manufacturing, where tactile sensing technology may impact the production process. By integrating tactile sensors into manufacturing systems, manufacturers gain precise, real-time feedback on product quality and assembly tasks. This advancement enables automation enhancement, defect detection, and optimization of product design and assembly processes, leading to reduced manufacturing costs and supply chain uncertainties. This I-Corps project is based on the development of smart sensing systems used in automatic manufacturing-compatible prototypes for high-resolution, conformal tactile sensor arrays. These arrays employ a piezoresistive sensing mechanism to measure forces across surfaces and seamlessly integrate them into wearable devices and robotic skins. Additionally, deep-learning techniques capable of extracting information from the sensor readouts have been devised. By analyzing the data captured by the sensing array, it is possible to determine user activity, state, and interactions with the environment. Notably, utilizing prototype tactile gloves and deep learning, successful identification and distinction of distinct patterns associated with human grasping have been achieved. This tactile sensing technology also has been incorporated into insoles, enabling the collection of data on the wearer's pose, activity, and environmental interactions. Furthermore, tactile sensing systems have been integrated into robot end-effectors and robotic skins that enable robots to perceive touch and perform complex manipulation tasks. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.