ABSTRACT
Public-private collaborative efforts to address healthcare challenges in low- and middle-income countries have been the focus of digital initiatives to improve both access and quality of health services. We report the early feasibility, experience, and learnings of migrating healthcare data generated from a proprietary, privately owned cloud-based environment into an on-premises National Health Data Center (NHDC) in compliance with Kenya's data management legislation. In 2018, Medtronic LABS entered into a partnership with the Kenya Ministry of Health and other stakeholders to improve access to quality services and data availability for non-communicable diseases (diabetes and hypertension), anchored on the SPICE digital health platform. Data migration from SPICE to the NHDC necessitated the establishment of multi-stakeholder coordination structures, alignment on system configuration requirements, provisioning of on-premises servers, data replication and monitoring. The data replication process showed consistency in format and content with no evidence of data loss. The monitoring of the server uptime and availability, however, exposed overall downtime of 15% of the total time tracked between April and December 2022 caused by Internet Protocol address configuration issues, power outages, firewall rule changes, and unscheduled system maintenance. Monthly tracked downtime however reduced from a high of 28% in April 2022 to 5% in December 2022. Our early experience shows that data migration from proprietary host environments to public "one-stop-shop" national data warehouses are feasible provided investments are made in the requisite infrastructure, software and human resource capacity to ensure long-term sustainability, maintenance, and scale to match cloud-based data hosting. Further, digital health solutions developed in collaboration with non-state actors can be integrated into national data systems, saving Governments the cost and efforts of building similar tools while leveraging private sector capacity.
ABSTRACT
Equipping soft robotic grippers with sensing and perception capabilities faces significant challenges due to their high compliance and flexibility, limiting their ability to successfully interact with the environment. In this work, we propose a sensorized soft robotic finger with embedded marker pattern that integrates a high-speed neuromorphic event-based camera to enable finger proprioception and exteroception. A learning-based approach involving a convolutional neural network is developed to process event-based heat maps and achieve specific sensing tasks. The feasibility of the sensing approach for proprioception is demonstrated by showing its ability to predict the two-dimensional deformation of three points located on the finger structure, whereas the exteroception capability is assessed in a slip detection task that can classify slip heat maps at a temporal resolution of 2 ms. Our results show that our proposed approach can enable complete sensorization of the finger for both proprioception and exteroception using a single camera without negatively affecting the finger compliance. Using such sensorized finger in robotic grippers may provide safe, adaptive, and precise grasping for handling a wide category of objects.
Subject(s)
Robotics , Fingers , Neural Networks, Computer , Proprioception , Hand StrengthABSTRACT
In recent years, robotic sorting is widely used in the industry, which is driven by necessity and opportunity. In this paper, a novel neuromorphic vision-based tactile sensing approach for robotic sorting application is proposed. This approach has low latency and low power consumption when compared to conventional vision-based tactile sensing techniques. Two Machine Learning (ML) methods, namely, Support Vector Machine (SVM) and Dynamic Time Warping-K Nearest Neighbor (DTW-KNN), are developed to classify material hardness, object size, and grasping force. An Event-Based Object Grasping (EBOG) experimental setup is developed to acquire datasets, where 243 experiments are produced to train the proposed classifiers. Based on predictions of the classifiers, objects can be automatically sorted. If the prediction accuracy is below a certain threshold, the gripper re-adjusts and re-grasps until reaching a proper grasp. The proposed ML method achieves good prediction accuracy, which shows the effectiveness and the applicability of the proposed approach. The experimental results show that the developed SVM model outperforms the DTW-KNN model in term of accuracy and efficiency for real time contact-level classification.
ABSTRACT
This paper presents the design and analysis of an intelligent control system that inherits the robust properties of sliding-mode control (SMC) for an n-link robot manipulator, including actuator dynamics in order to achieve a high-precision position tracking with a firm robustness. First, the coupled higher order dynamic model of an n-link robot manipulator is briefy introduced. Then, a conventional SMC scheme is developed for the joint position tracking of robot manipulators. Moreover, a fuzzy-neural-network inherited SMC (FNNISMC) scheme is proposed to relax the requirement of detailed system information and deal with chattering control efforts in the SMC system. In the FNNISMC strategy, the FNN framework is designed to mimic the SMC law, and adaptive tuning algorithms for network parameters are derived in the sense of projection algorithm and Lyapunov stability theorem to ensure the network convergence as well as stable control performance. Numerical simulations and experimental results of a two-link robot manipulator actuated by DC servo motors are provided to justify the claims of the proposed FNNISMC system, and the superiority of the proposed FNNISMC scheme is also evaluated by quantitative comparison with previous intelligent control schemes.
