Narrow Results

Agriculture catalyzes the economy in developing nations. Malaysian agriculture constitutes 4.06 million hectares, with 80% encompassing industrial crops and agro-food production, boosting the economy through implementing precision agriculture (PA). Precision agriculture gives minimal environmental implications by using an unmanned aerial vehicle (UAV), improving sustainability, productivity, and crop production 30-fold instead of conventional methods. This study aims to review the UAV application based on technical requirements with insights into the potentiality of precision agriculture in UAV agriculture technologies, limitations, and solutions.

With a tremendous increase in mobile and wearable devices, the study of sensor-based activity recognition has drawn a lot of attention in the past years. In recent years, the applications of Human Activity Recognition are getting more and more attention, especially in eldercare and healthcare as an assistive technology when combined with the Internet of Things. In this paper, we propose three deep learning approaches to improve the accuracy of activity detection on the WISDM dataset. Particularly, we apply a convolutional neural network to extract the interesting features, then we use softmax function, support vector machine, and random forest for classification tasks. The results show that the hybrid algorithm, convolutional neural network combined with the support vector machine, outperforms all the previous methods in classifying every activity. In addition, not only the support vector machine but also the random forest shows better accuracy in classification task than the neural network classification and the former approaches do.

We propose a new Dynamic Address Assignment Protocol for wireless sensor networks. This new protocol is based on the Dynamic Host Configuration Protocol and is modified into wireless sensor networks. The protocol aims to increase the flexibility and agility of wireless sensor networks with changing the amount of sensor nodes in sensor networks system easily. It is well suit for large-scale wireless sensor networks which update their sensor nodes frequently. The goals are achieved by assigning address dynamically and registering in collectors. We present an implement on wireless sensor network and evaluate our approach by discussing the payload ability of a wireless node.

This paper develops a cable-less pipeline robot which can adapt to the change of pipe diameter. It introduces the composition and function of robot mechanism as well as control system in detail. In this study, the robot employs the crawler traveling mechanism with symmetrical distribution in three groups. Through the rotating of ball screw and the change of the quadrilateral structure both driven by the stepper motor, the radial dimension of robot will be changed. The control system of pipeline robot adopts the modularized design method, takes stm32f103 as its embedded processor, provides a variety of hardware functions as well as the expansion of the circuit interfaces, has a high versatility and applicability. Through verification, the designed pipeline robot can work in various environments of pipeline, realizing in-pipe detection, data acquisition as well as other functions.

The paper determines the optimal position of thermistors in sensitive components of gas pendulum omnibearing level posture sensor is d=900 μm (d is distance of two relatively detection thermistors). By using the finite element method, through the establishment of three-dimensional modeling of the thermistors of sensitive element in different positions, different tilting states, the distribution of temperature field and flow field inside the sensitive element are calculated. The results show that: (1) When the d is changed, the temperature field and flow field changes. (2) The relation curve, which is about the difference Δv of airflow velocity around the two relatively detection thermistors and tilt angle, changes with the d. This opens the effective approaches for the optimization design of the sensor.

The Internet of Things is a new type of technical system combining various types of information technology, the application of Internet of Things in agriculture meets the requirement of modern agriculture development and represents the orientation of future agriculture. In this article the Internet of Things system in agriculture is analyzed according to its systematic structure including its perception level, transmission level and application level and perception level, the sensor is the core of this level, multidimensional agricultural information can be obtained by advanced technology such as sensing and remote sensing about crops, soil and environment etc. The transmission level consists of wireless transmission network combing wireless transmission methods such as Zigbee, GPRS, WIFI and Bluetooth and network technology. The application level involves in intelligent management of agriculture including multi-dimensional information integration, intelligent decision and automatic control. The orientation of research and development is shown for the Internet of Things system in agriculture in this article. Its development will bring a brand-new reform for modern agriculture and realize agricultural automation and intellectualization.