In video streaming services, predicting the continuous user’s quality of experience (QoE) plays a crucial role in delivering high quality streaming contents to the user. However, the complexity caused by the temporal dependencies in QoE data and the non-linear relationships among QoE influence factors has introduced challenges to continuous QoE prediction. To deal with that, existing studies have utilized the Long Short-Term Memory model (LSTM) to effectively capture such complex dependencies, resulting in excellent QoE prediction accuracy. However, the high computational complexity of LSTM, caused by the sequential processing characteristic in its architecture, raises a serious question about its performance on devices with limited computational power. Meanwhile, Temporal Convolutional Network (TCN), a variation of convolutional neural networks, has recently been proposed for sequence modeling tasks (e.g., speech enhancement), providing a superior prediction performance over baseline methods including LSTM in terms of prediction accuracy and computational complexity. Being inspired of that, in this paper, an improved TCN-based model, namely CNN-QoE, is proposed for continuously predicting the QoE, which poses characteristics of sequential data. The proposed model leverages the advantages of TCN to overcome the computational complexity drawbacks of LSTM-based QoE models, while at the same time introducing the improvements to its architecture to improve QoE prediction accuracy. Based on a comprehensive evaluation, we demonstrate that the proposed CNN-QoE model can provide a high QoE prediction performance on both personal computers and mobile devices, outperforming the existing approaches.

Computer vision plays a significant role in mobile robot navigation due to the wealth of information extracted from digital images. Mobile robots localize and move to the intended destination based on the captured images. Due to the complexity of the environment, obstacle avoidance still requires a complex sensor system with a high computational efficiency requirement. This study offers a real-time solution to the problem of extracting corridor scenes from a single image using a lightweight semantic segmentation model integrating with the quantization technique to reduce the numerous training parameters and computational costs. The proposed model consists of an FCN as the decoder and MobilenetV2 as the decoder (with multi-scale fusion). This combination allows us to significantly minimize computation time while achieving high precision. Moreover, in this study, we also propose to use the Balance Cross-Entropy loss function to handle diverse datasets, especially those with class imbalances and to integrate a number of techniques, for example, the Adam optimizer and Gaussian filters, to enhance segmentation performance. The results demonstrate that our model can outperform baselines across different datasets. Moreover, when being applied to practical experiments with a real mobile robot, the proposed model’s performance is still consistent, supporting the optimal path planning, allowing the mobile robot to efficiently and effectively avoid the obstacles.

Accurate detection and estimation of pallet poses from color and depth data (RGB-D) are integral components many in advanced warehouse intelligent systems. State-of-the art object pose estimation methods follow a two-stage process, relying on off-the-shelf segmentation or object detection in the initial stage and subsequently predicting the pose of objects using cropped images. The cropped patches may include both the target object and irrelevant information, such as background or other objects, leading to challenges in handling pallets in warehouse settings with heavy occlusions from loaded objects. In this study, we propose an innovative deep learning-based approach to address the occlusion problem in pallet pose estimation from RGB-D images. Inspired by the selective attention mechanism in human perception, our developed model learns to identify and attenuate the significance of features in occluded regions, focusing on the visible and informative areas for accurate pose estimation. Instead of directly estimating pallet poses from cropped patches as in existing methods, we introduce two feature map re-weighting modules with cross-modal attention. These modules effectively filter out features from occluded regions and background, enhancing pose estimation accuracy. Furthermore, we introduce a large-scale annotated pallet dataset specifically designed to capture occlusion scenarios in warehouse environments, facilitating comprehensive training and evaluation. Experimental results on the newly collected pallet dataset show that our proposed method increases accuracy by 13.5% compared to state-of-the-art methods.

This paper addresses the formidable challenge of enabling autonomous navigation in Mobile Robots (MRs), focusing on the development of advanced path planning strategies. Despite their pivotal role in diverse applications, they face challenges in dynamic settings due to limitation in existing Global Path Planning (GPP) and Local Path Planning (LPP) techniques. In response to this, we propose an innovative hybrid path planning approach that enhances the A* algorithm with a risk-aware heuristic function and integrates the Jump Point Search (JPS) technique for route optimization. Additionally, B-spline smoothing is employed for perceptually global trajectory refinement. Our approach also includes an innovative improvement to the Dynamic Window Approach (DWA) to align with the proposed enhanced A* algorithm for effective local navigation. Acknowledging the importance of high-quality input in path planning, we present substantial improvements to the IRDC-Net, a monocular-image semantic-segmentation model that we studied previously. Novel improvements include the integration of quantization and the Adam optimizer, along with the implementation of the Balanced Cross-Entropy loss function. These enhancements not only elevate the output quality of IRDC-Net but also reduce the model’s training parameters. The experimental results demonstrate the performance and viability of the proposed algorithm. Ultimately, the hybrid MR’s path planning algorithm exhibits proficiency across various tasks, particularly in addressing the challenge of evading moving obstacles to ensure the robot’s safety while adhering to the global path.

In this paper, we present a vision-based pick-and-place control system for industrial robots using an eye-in-hand camera. In industry, using robots with cameras greatly improves efficiency and performance. Previous studies have focused on utilizing robotic arms for the pick-and-place process in simulated environments. The challenge when experimenting with real systems lies in aligning the coordinate systems between the robot and the camera, as well as ensuring high data accuracy during experimentation. To address this issue, our research focuses on utilizing a low-cost 2D camera combined with deep learning algorithms mounted on the end-effector of the robotic arm. This study is evaluated in both simulation and real-world experiments. We propose a novel approach that combines the YOLOv7 (You Only Look Once V7) deep learning network with GAN (Generative Adversarial Networks) to achieve fast and accurate object recognition. This system uses deep learning to process camera data to extract object positions for the robot in real-time. Due to its advantages of fast inference and high accuracy, YOLO is applied as the baseline for research. By training the deep learning model on diverse objects, it effectively recognizes and detects any object in the robot’s workspace. Through experimental results, we demonstrate the feasibility and effectiveness of our vision-based pick-and-place system. Our research contributes an important advancement in the field of industrial robots by showcasing the potential of using a 2D camera and an integrated deep learning system for object manipulation.

Adaptive Educational Hypermedia Systems (AEHS) play a crucial role in supporting adaptive learning and immensely outperform learner-control based systems. AEHS’ page indexing and hyperspace rely mostly on navigation supports which provide the learners with a user-friendly interactive learning environment. Such AEHS features provide the systems with a unique ability to adapt learners’ preferences. However, obtaining timely and accurate information for their adaptive decision-making process is still a challenge due to the dynamic understanding of individual learner. This causes a spontaneous changing of learners’ learning styles that makes hard for system developers to integrate learning objects with learning styles on real-time basis. Thus, in previous research studies, multiple levels navigation supports have been applied to solve this problem. However, this approach destroys their learning motivation because of imposing time and work overload on learners. To address such a challenge, this study proposes a bioinformatics-based adaptive navigation support that was initiated by the alternation of learners’ motivation states on a real-time basis. EyeTracking sensor and adaptive time-locked Learning Objects (LOs) were used. Hence, learners’ pupil size dilation and reading and reaction time were used for the adaption process and evaluation. The results show that the proposed approach improved the AEHS adaptive process and increased learners’ performance up to 78%.