Born from the machine learning (ML) subfield of neural networks (NN), deep learning (DL) has many advantages over other ML algorithms and has become more significant today. As one of the most essential model architectures of DL, the convolutional neural network (CNN) has attracted the attention of many researchers, especially in recent years. Meanwhile, sentiment analysis has become more renowned since the rapid development of various online platforms like blogs, social networks, etc. To study these two heated topics together, this article selects a particular CNN model designed for sentiment analysis and explores its width’s potential influence on the result. During the experiment, four CNN models are created based on the same structure but with increasing width. By forwarding the pre-processed datasets to the four models and comparing their performances from different perspectives using different metrics, it’s concluded that the more expansive the model's width, the better it performs in the training, validation, and testing sections.

Show more

With the progress of science and technology, more and more cars have the function of automatic driving. After reviewing a series of articles, it is found that there are still many problems in the research on automatic driving. These include problems with sensors, cameras, navigation systems, signals and so on. The primary focus of this article is the automobile's sensor system, which includes body-sensing sensors, radar sensors, vision sensors, and GPS systems. However, there will be numerous issues if these sensors are used alone. It can be seen that many researchers have done the application of multiple sensors and achieved good results. Thus, in order to enable the widespread adoption of automated driving in the future, it is advised to integrate two or three sensors and conduct testing through real-world applications. This can not only reduce the occurrence of accidents but also promote the broader development of autonomous driving.

Show more

This paper compares three Simultaneous Localization and Mapping (SLAM) algorithms. SLAM algorithms are the core technology for autonomous navigation and environmental perception of mobile robots. SLAM algorithms are used by mobile robots to perceive the surrounding environment, build up an environment map and position themselves in real-time in an unknown environment. This article first systematically reviews the basic principles of each algorithm based on experiments and studies that have been completed by previous researchers and illustrates their respective unique mechanisms for processing sensor data, map construction, and localization. Subsequently, this paper analyzes the performance differences and characteristics of the three algorithms in practical applications in terms of robustness in complex environments, consumption of computing resources, and accuracy for generated maps. Finally, based on the advantages and disadvantages of each analyzed algorithm, this article summarizes the most suitable and unsuitable usage scenarios of different algorithms in specific situations. Moreover, this article puts forward specific algorithm selection suggestions for different scenarios to help engineers and researchers make more appropriate decisions in actual projects.

Show more

This paper examines the perception systems used in industrial Automated Guided Vehicles (AGVs), focusing on traditional and advanced sensor solutions. Traditional perception methods, such as track-based and magnetic tape guidance, offer reliability but are limited in flexibility. In contrast, radar, vision, and LiDAR sensors provide enhanced perception capabilities, enabling AGVs to navigate safely and efficiently in complex industrial environments. The study explores various sensors, including visible light, infrared, ultrasonic, LiDAR, magnetic strip sensors, Inertial Measurement Units (IMUs), tactile sensors, Ultrawideband (UWB) sensors, thermal sensors, and millimeter-wave (mmWave) sensors, highlighting their principles, advantages, and limitations. The integration of these sensors supports robust navigation and operational efficiency in diverse settings. The methodology involves reviewing existing literature and analyzing current technologies used in industrial AGVs. Results indicate that while traditional solutions are reliable, advanced sensor technologies significantly enhance AGV performance. The paper concludes that the future of AGV perception systems lies in the integration of advanced sensors with artificial intelligence and machine learning algorithms, promoting intelligent and adaptive industrial automation. Additionally, it underscores the necessity of developing robust sensor fusion techniques to harness the full potential of these advanced sensors.

Show more