Article
Automation & Control Systems
Yan Meng, Zhengxing Wu, Pengfei Zhang, Jian Wang, Junzhi Yu
Summary: This article proposes a novel estimation-and-prediction framework for real-time digital video stabilization of bioinspired robotic fish. By establishing a camera-IMU model and using a translation estimation network, it effectively reduces the instability of the robot's camera path. Experimental results demonstrate that this method is more effective in maintaining visual stability and improving stabilization speed compared to other methods.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Computer Science, Software Engineering
Mathieu Faverge, Nathalie Furmento, Abdou Guermouche, Gwenole Lucas, Raymond Namyst, Samuel Thibault, Pierre-Andre Wacrenier
Summary: Task-based systems are popular for utilizing the computational power of heterogeneous systems. We have extended the Sequential Task Flow (STF) model with hierarchical tasks in the StarPU runtime system to create a more dynamic task graph and adjust granularity at runtime. This solves the issues of submission overhead and static task graphs not being well-suited for heterogeneous systems.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Computer Science, Information Systems
Eduardo Guerra, Antonio De Oliveira Dias, Luiz Gustavo D. O. Veras, Ademar Aguiar, Joelma Choma, Tiago Silva Da Silva
Summary: The Adaptive Object Model (AOM) is an architectural style that allows domain entity types to be changed at runtime for higher flexibility. This study proposes a model to reuse frameworks designed for classic object-oriented domain models in an AOM application, by using dynamically-generated adapters for AOM entities. The research shows that the proposed model can successfully be employed to use AOM entities with frameworks that were not originally designed for AOM applications, reducing the effort needed to adopt an AOM architecture.
Article
Computer Science, Interdisciplinary Applications
Hamid Majidi Balanji, Ali Emre Turgut, Lutfi Taner Tunc
Summary: The demand for high quality industrial robots with good repeatability and accuracy has increased. Calibration using tracking devices is a practical approach to sustain accuracy, with different devices like laser trackers and stereo cameras utilized. A novel calibration framework based on computer vision techniques using ArUco markers and a single camera has been proposed for kinematic modeling of robots, showing promising results in real world scenarios.
ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING
(2022)
Article
Computer Science, Theory & Methods
Lazaros Papadopoulos, Dimitrios Soudris, Christoph Kessler, August Ernstsson, Johan Ahlqvist, Nikos Vasilas, Athanasios Papadopoulos, Panos Seferlis, Charles Prouveur, Matthieu Haefele, Samuel Thibault, Athanasios Salamanis, Theodoros Ioakimidis, Dionysios Kehagias
Summary: Programming upcoming exascale computing systems is challenging, but the EXA2PRO framework improves developers' productivity by encapsulating low-level optimizations and supporting diverse computing architectures.
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
(2022)
Article
Computer Science, Information Systems
A. Romero-Garces, R. Salles De Freitas, R. Marfil, C. Vicente-Chicote, J. Martinez, J. F. Ingles-Romero, A. Bandera
Summary: This paper proposes an approach to self-adaptation in robots by modeling behavior variability at design-time and allowing the robot to configure its behavior at runtime based on contextual information. This approach is supported by a model-based framework that allows robotic engineers to specify behavior variation points, contextual information, and non-functional properties for measuring Quality-of-Service (QoS) of the robot. The framework automatically generates the runtime infrastructure for the robot to adapt its behavior and achieve the best QoS according to its current context.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Zhuming Bi, Zhonghua Miao, Bing Zhang, Chris W. J. Zhang
Summary: In a dynamic manufacturing environment, a system should be reconfigurable to meet the demands of mass customization. However, the lack of scientific methods and tools to support system reconfiguration poses a challenge for designers in adapting to changing needs.
IEEE SYSTEMS JOURNAL
(2021)
Article
Computer Science, Hardware & Architecture
Haikun Liu, Jiahong Xu, Xiaofei Liao, Hai Jin, Yu Zhang, Fubing Mao
Summary: In this article, the authors propose a simulation framework called MHSim to evaluate the energy efficiency and performance of applications running with memristor-based accelerators (MBA) and CPUs. They design a general-purpose MBA for floating-point computation models in matrix-matrix multiplication. The experimental results show that the deviations of energy consumption and latency are minimal compared to SPICE-based simulation.
IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
(2022)
Article
Robotics
Santeri Lampinen, Longchuan Niu, Lionel Hulttinen, Jouni Niemi, Jouni Mattila
Summary: In the mining industry, crushing of blasted ore is crucial for extracting valuable minerals, typically done in multiple stages where the first stage's performance depends on the ore's size distribution. To automate the boulder breaking tasks, an intelligent robotic control system and visual perception system are essential.
JOURNAL OF FIELD ROBOTICS
(2021)
Article
Robotics
Toshiki Fujishiro, Tadayoshi Aoyama, Kazuki Hano, Masaki Takasu, Masaru Takeuchi, Yasuhisa Hasegawa
Summary: The study proposed a microinjection system that improves depth visibility through real-time 3D image presentation, estimating 3D position and size of micro-targets using a focal position adjustment mechanism. The system also developed a calibration method to adjust relative position and orientation between a target object and micromanipulators, presenting reconstructed 3D images of a target and micromanipulators for manipulation experiments involving porcine embryos.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Automation & Control Systems
Yongxiang Wu, Fuhai Zhang, Yili Fu
Summary: This article introduces a novel anchor-free grasp detector based on a fully convolutional network for real-time detection of multiple valid grasps from RGB-D images. By directly predicting grasps at feature points, the method eliminates predefined anchors, improving computational efficiency and accuracy.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Computer Science, Hardware & Architecture
Francesco Lumpp, Stefano Aldegheri, Hiren D. Patel, Nicola Bombieri
Summary: This article addresses the performance challenge of OpenVX on heterogeneous multi-core platforms by implementing static task scheduling and mapping using HEFT and XEFT algorithms, achieving significant performance improvements that are promising for applications in edge computing such as intelligent video analytics.
IEEE TRANSACTIONS ON COMPUTERS
(2021)
Article
Computer Science, Artificial Intelligence
R. Udendhran, G. Yamini, N. Badrinath, J. Jegathesh Amalraj, A. Suresh
Summary: Robot vision, enabled by machine vision systems and deep learning, allows robots to operate faster in various contexts. This technology is driven by advancements in machine vision systems, which consist of programs, cameras, and other technologies that assist robots in developing visual insights.
Article
Engineering, Electrical & Electronic
Yu Wang, Chong Tang, Mingxue Cai, Jiye Yin, Shuo Wang, Long Cheng, Rui Wang, Min Tan
Summary: This study presents a real-time underwater onboard vision sensing system for robotic gripping, including image enhancement method, real-time lightweight object detector design and deployment method, and refraction tracing model establishment. Experimental results demonstrate the practical value of these methods in underwater robotic gripping tasks.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Artificial Intelligence
Andres Ussa, Chockalingam Senthil Rajen, Tarun Pulluri, Deepak Singla, Jyotibdha Acharya, Gideon Fu Chuanrong, Arindam Basu, Bharath Ramesh
Summary: This article proposes a real-time, hybrid neuromorphic framework for object tracking and classification using event-based cameras. It addresses the challenge of deep learning inference on low-power, embedded platforms. A hardware-friendly object tracking scheme is implemented using a frame-based region proposal method and the frame-based object track input is converted back to spikes for classification. The proposed neuromorphic system is also compared to state-of-the-art methods and demonstrated in real-time and embedded applications.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
