Shengyu Fang

Abstract Multi-sensor data fusion is to take full advantage of the complementary nature of multivariate data to improve the feasibility of the statistics. The weighted fusion algorithm is commonly used due to its easiness to achieve. Among the relative factors, the weight directly impacts the results of the data fusion, therefore, the selection of weight is particularly important, as choosing an inappropriate weight will lead to the instability of algorithm performance. To find the best weight, we develop an improved weighted fusion algorithm, introducing the concept of the optimal proportion weight and using secondary weighted approach ----- the single sensor will be weighted individually before the whole sensor system is weighted in order to achieve the optimal algorithm performance.

Abstract This article designs the entire intelligent Greenhouse system architecture, collects a large amount of data such as temperature, humidity, light, and water through Multi-sensors, and implements an improved fuzzy neural network control algorithm. The advantage of this algorithm over traditional neural networks is that it can express fuzzy and qualitative the knowledge,which also improves the overall learning ability of the system. This intelligent system combines advanced technologies such as AIoT, NB-IoT, and 5G and uploads sensor data to the cloud platform. Through this improved learning algorithm, it provides optimized temperature, humidity and lighting conditions for plants in the greenhouse, and optimal growth conditions. In addition, this system design also optimizes NB-IoT power management, which can improve the overall system information transmission rate and decoding efficiency, and better achieve connection between the 5G and NB-IoT in order to link up each operations and for transmissions. That design and research of the intelligent greenhouse control system has improved the function and the value of the traditional greenhouse control system. Through the AIoT fusion technology, the entire intelligent greenhouse system is made more convenient, intelligent and integrated.

Abstract UAV flight control systems have the characteristics of non-linear, multi-variable, and strong coupling. Traditional PID control algorithms have many problems in complex flight environments, such as large adjustable parameters, poor robustness, and slow convergencein. Due to the defects of traditional PID control algorithm, this paper proposes an improved PID control algorithm based on fuzzy neurons. The modified fuzzy neuron is used to modify the traditional PID control algorithm. Through Matlab experimental simulation, it is proved that the algorithm has a significant improvement in response speed, robustness, accuracy and anti-interference compared with the traditional PID algorithm.

A pedestrians counting system based on head feature is established,which is suitable for indoor scene with cameras vertically installed.Firstly,the head model is established to detect the targets,and the model parameters are determined by the feature of shape and color.Secondly,a tracking algorithm is proposed to analyze the targets trajectory through template matching and state prediction.The experimental results show that the system has good effect to bidirectional pedestrians counting,and the accuracy is over 95%.