This paper presents a framework addressing the challenge of global localization in autonomous mobile robotics by integrating LiDAR-based descriptors and Wi-Fi finger-printing in a pre-mapped environment. This is motiv...
This paper presents a framework addressing the challenge of global localization in autonomous mobile robotics by integrating LiDAR-based descriptors and Wi-Fi finger-printing in a pre-mapped environment. This is motivated by the increasing demand for reliable localization in complex scenarios, such as urban areas or underground mines, requiring robust systems able to overcome limitations faced by traditional Global Navigation Satellite System (GNSS)-based localization methods. By leveraging the complementary strengths of LiDAR and Wi-Fi sensors used to generate predictions and evaluate the confidence of each prediction as an indicator of potential degradation, we propose a redundancy-based approach that enhances the system's overall robustness and accuracy. The proposed framework allows independent operation of the LiDAR and Wi-Fi sensors, ensuring system redundancy. By combining the predictions while considering their confidence levels, we achieve enhanced and consistent performance in localization tasks.
‘KnowSeq’ R Package includes all the essential tools to carry out transcriptomic analysis, providing intuitive functions to build efficient and robust pipelines. In this paper, its capacities are demonstrated in a p...
详细信息
This paper describes an academic project building on various state-of-the-art Industry 4.0 technologies. The project considers as the primary scenario the current problem behind the accumulation of biological waste in...
详细信息
ISBN:
(数字)9781665421614
ISBN:
(纸本)9781665421621
This paper describes an academic project building on various state-of-the-art Industry 4.0 technologies. The project considers as the primary scenario the current problem behind the accumulation of biological waste in the vicinity of hospitals due to the COVID-19. To that end, a fully automated personal protective equipment recycling plant is developed using the Factory I/O simulation tool, TIA Portal, and Ignition. The project is considered to serve well as a reference for other Electrical Engineering undergraduates while complementing their training in skills such as design, automation, and supervision of industrial processes.
The performance of neural networks has granted deep learning a place at the forefront of machine learning in the last decade. Although these models are computationally intensive, their advantage is recognized in a wid...
详细信息
In magnetic resonant coupling (MRC) based wireless power transfer (WPT) systems, receiver (RX) feedback communication is promising to enhance the capability and efficiency of the system. Although some studies have exp...
详细信息
The old city of Mosul possesses a rich built heritage, featuring distinct architectural styles evolved, incorporating indigenous knowledge and expertise to address functional, climatic, economic, and construction requ...
详细信息
Entity relation extraction consists of two sub-tasks: entity recognition and relation extraction. Existing methods either tackle these two tasks separately or unify them with word-byword interactions. In this paper, w...
详细信息
Dielectric Elastomer Actuators (DEAs) can be seen as flexible capacitors, where an incompressible elastomer is sandwiched between two compliant electrodes. From the elastomer non-linear and time-dependent phenomena ar...
Dielectric Elastomer Actuators (DEAs) can be seen as flexible capacitors, where an incompressible elastomer is sandwiched between two compliant electrodes. From the elastomer non-linear and time-dependent phenomena arise during the activation of the actuator. This results in the need for appropriate control methods, especially when they are used in the field of soft robots. Therefore, this paper presents a feedforward controller which is based on Spiking Neural Networks (SNNs). To the best of our knowledge this is the first time DEAs and SNNs are combined to make a step towards true soft robots. It was demonstrated that it is necessary to include feedback into the control scheme. Also, it was found that in the context of a SpiNN-3 development board as a neuromorphic hardware platform, neuron encoding is the most promising encoding method. This also implies that a classification approach is preferable to a regression approach.
The large increase in the number of Internet of Things (IoT) devices have revolutionised the way data is processed, which added to the current trend from cloud to edge computing has resulted in the need for efficient ...
详细信息
Detecting and classifying vehicles is a modern technology with numerous uses. Administration and regulation of traffic is one of the primary uses. Projects utilizing image processing to prevent traffic accidents heavi...
详细信息
ISBN:
(数字)9798350372120
ISBN:
(纸本)9798350372137
Detecting and classifying vehicles is a modern technology with numerous uses. Administration and regulation of traffic is one of the primary uses. Projects utilizing image processing to prevent traffic accidents heavily rely on vehicle monitoring and detection. Monitoring and recording human movement in surveillance situations require the ability to follow moving objects. Considering its significance, it provides a valuable image processing-based vehicle detection technique—a vehicle tracking and detection system based on images. The recently released high-resolution road vehicle dataset supports deep learning-based vehicle detection and monitoring on the Python platform. It includes over 100 well-defined pictures taken from movies from various locations. This data is produced using a range of image processing methods. The most recent iteration of the YOLO model, the v5 model, was employed for detection. R-CNN is additionally utilized for model detection. The four stages in the Region-Based Convolutional Neural Network (RCNN) process are preprocessing, segmentation, feature selection, and classification. The first step is identifying moving vehicles accurately. Preprocessing the original image is necessary. This enhances the image’s quality, valuable information extracted, or unwanted areas cropped out. Segmentation is the second step. The region’s borders are established, and the entire image is divided into numerous smaller areas using a threshold. The vehicle image’s most pertinent aspects or characteristics (features) that aid in precise detection are found in the third stage, feature selection. A region-based convolutional neural network (RCNN) is employed to determine the categorization. The Yolov5 method can be used to identify and categorize items. The Region-based Convolutional Neural Network (RCNN) approach enhances accuracy and temporal complexity over current segmentation algorithms.
暂无评论