Search for: All records

This paper presents an attention-based, deep learning framework that converts robot camera frames with dynamic content into static frames to more easily apply simultaneous localization and mapping (SLAM) algorithms. The vast majority of SLAM methods have difficulty in the presence of dynamic objects appearing in the environment and occluding the area being captured by the camera. Despite past attempts to deal with dynamic objects, challenges remain to reconstruct large, occluded areas with complex backgrounds. Our proposed Dynamic-GAN framework employs a generative adversarial network to remove dynamic objects from a scene and inpaint a static image free of dynamic objects. The Dynamic-GAN framework utilizes spatial-temporal transformers, and a novel spatial-temporal loss function. The evaluation of Dynamic-GAN was comprehensively conducted both quantitatively and qualitatively by testing it on benchmark datasets, and on a mobile robot in indoor navigation environments. As people appeared dynamically in close proximity to the robot, results showed that large, feature-rich occluded areas can be accurately reconstructed with our attention-based deep learning framework for dynamic object removal. Through experiments we demonstrate that our proposed algorithm has up to 25% better performance on average as compared to the standard benchmark algorithms. 

Microassembly systems utilizing precision robotics have long been used for realizing 3-dimensional microstructures such as microrobots. Prior to assembly, such components are fabricated using Micro-Electro-Mechanical-System (MEMS) technology. The microassembly system then directs a microgripper through automated or human-controlled pick-and-place operations. In this paper, we describe a novel custom microassembly system, named NEXUS. The NEXUS integrates multi-degree of freedom (DOF) precision positioners, microscope computer vision, and micro-scale process tools such as a microgripper and vacuum tip. A semi-autonomous human-machine interface (HMI) is programmed by NI LabVIEW® to allow the operator to interact with the microassembly system. The NEXUS human-machine interface includes multiple functions, such as positioning, target detection, visual servoing, and inspection. The microassembly system’s HMI was used by operators to assemble various 3-dimensional microrobots such as the Solarpede, a novel light-powered stick-and-slip mobile microcrawler. Experimental results are reported in this paper that evaluate the system’s semi-autonomous capabilities in terms of assembly rate and yield and compare them to purely teleoperated assembly performance. Results show that the semi-automated capabilities of the microassembly system’s HMI offer a more consistent assembly rate of microrobot components.