Abstract:Stiffness estimation is crucial for delicate object manipulation in robotic and prosthetic hands but remains challenging due to dependence on force and displacement measurement and real-time sensory integration. This study presents a piezoelectric sensing framework for stiffness estimation at first contact during pinch grasps, addressing the limitations of traditional force-based methods. Inspired by human skin, a multimodal tactile sensor that captures vibrational and force data is developed and integrated into a prosthetic hand's fingertip. Machine learning models, including support vector machines and convolutional neural networks, demonstrate that vibrational signals within the critical 15 ms after first contact reliably encode stiffness, achieving classification accuracies up to 98.6\% and regression errors as low as 2.39 Shore A on real-world objects of varying stiffness. Inference times of less than 1.5 ms are significantly faster than the average grasp closure time (16.65 ms in our dataset), enabling real-time stiffness estimation before the object is fully grasped. By leveraging the transient asymmetry in grasp dynamics, where one finger contacts the object before the others, this method enables early grasp modulation, enhancing safety and intuitiveness in prosthetic hands while offering broad applications in robotics.
Abstract:This study presents a proof of concept for a contactless elevator operation system aimed at minimizing human intervention while enhancing safety, intelligence, and efficiency. A microcontroller-based edge device executing tiny Machine Learning (tinyML) inferences is developed for elevator operation. Using person detection and keyword spotting algorithms, the system offers cost-effective and robust units requiring minimal infrastructural changes. The design incorporates preprocessing steps and quantized convolutional neural networks in a multitenant framework to optimize accuracy and response time. Results show a person detection accuracy of 83.34% and keyword spotting efficacy of 80.5%, with an overall latency under 5 seconds, indicating effectiveness in real-world scenarios. Unlike current high-cost and inconsistent contactless technologies, this system leverages tinyML to provide a cost-effective, reliable, and scalable solution, enhancing user safety and operational efficiency without significant infrastructural changes. The study highlights promising results, though further exploration is needed for scalability and integration with existing systems. The demonstrated energy efficiency, simplicity, and safety benefits suggest that tinyML adoption could revolutionize elevator systems, serving as a model for future technological advancements. This technology could significantly impact public health and convenience in multi-floor buildings by reducing physical contact and improving operational efficiency, particularly relevant in the context of pandemics or hygiene concerns.
Abstract:Skull Stripping is a requisite preliminary step in most diagnostic neuroimaging applications. Manual Skull Stripping methods define the gold standard for the domain but are time-consuming and challenging to integrate into pro-cessing pipelines with a high number of data samples. Automated methods are an active area of research for head MRI segmentation, especially deep learning methods such as U-Net architecture implementations. This study compares Vanilla, Residual, and Dense 2D U-Net architectures for Skull Stripping. The Dense 2D U-Net architecture outperforms the Vanilla and Residual counterparts by achieving an accuracy of 99.75% on a test dataset. It is observed that dense interconnections in a U-Net encourage feature reuse across layers of the architecture and allow for shallower models with the strengths of a deeper network.