Driving Activity Classification Using Deep Residual Networks Based on Smart Glasses Sensors

Accidents are still an issue in an intelligent transportation system,despite developments in self-driving technology(ITS).Drivers who engage in risky behavior account for more than half of all road accidents.As a result,reckless driving behaviour can cause congestion and delays.Computer vision and multimodal sensors have been used to study driving behaviour categorization to lessen this problem.Previous research has also collected and analyzed a wide range of data,including electroencephalography(EEG),electrooculography(EOG),and photographs of the driver’s face.On the other hand,driving a car is a complicated action that requires a wide range of body move-ments.In this work,we proposed a ResNet-SE model,an efficient deep learning classifier for driving activity clas-sification based on signal data obtained in real-world traffic conditions using smart glasses.End-to-end learning can be achieved by combining residual networks and channel attention approaches into a single learning model.Sensor data from 3-point EOG electrodes,tri-axial accelerometer,and tri-axial gyroscope from the Smart Glasses dataset was utilized in this study.We performed various experiments and compared the proposed model to base-line deep learning algorithms(CNNs and LSTMs)to demonstrate its performance.According to the research results,the proposed model outperforms the previous deep learning models in this domain with an accuracy of 99.17%and an F1-score of 98.96%.

Impacts of Human Activity on Biodiversity——An Approach to Driving Force Indicators

The impact of human activity on biodiversity is very complicated, making it difficult to design practical indicators for assessment.Currently,state and response indicators are listed within Chapter15 of Agenda21,but no mention is made of driving force indicators and how they relate to biodiversity protection. This paper repre- sents an effort to identify and then operationalize the idea of driving force indicators at national level.Five human induced factors affecting biodiversity were discussed: (a) habitat loss and fragmentation; (b) overexploitation of resources;(c)species introduction;(d)pollution;and (e) climate change. From these five factors, a subsetwas selected to serve as possible driving force indicators:(1)habitat loss, (2)the ratio of exotic species to indigenous ones, and (3)the change in pollution status. Although the three indicators are relatively simple,they coverthe most important human impacts on biodiversity and offer the potential for further redefinition and ultimate use within the spirit of biodiversity protection.

Study on Active Drive System for Colonoscope

The paper deals with active drive system for colonoscope. The system is mainly composed of soft mobile mechanism for earthworm locomotion and turning mechanism based on shape memory effect. The soft mobile mechanism contacts colon wall with air in inflatable balloons, so the robot has better soft and non invasive properties. The turning mechanism can be actively bent by shape memory alloy components. It ensures the colonoscope to adapt to the tortuous shape of colon. Some experiment results are given in the paper.

Passive Matrix Organic Electroluminescent Display for 3G Cellular Phone

The major features of the display technologies for 3G cellular phone are reviewed. The comparison between their potential candidates for 3G cellular phone is given, and a detailed discussion is made on passive matrix organic electroluminescent display technology. A novel method to improve display contrast ratio is presented. Finally several 3G phone set prototypes with OLED display panels are given as well as the market forecast.

A driving pulse edge modulation technique and its complex programming logic devices implementation

With the continual increase in switching speed and rating of power semiconductors, the switching voltage spike becomes a serious problem. This paper describes a new technique of driving pulse edge modulation for insulated gate bipolar transistors(IGBTs). By modulating the density and width of the pulse trains, without regulating the hardware circuit, the slope of the gate driving voltage is controlled to change the switching speed. This technique is used in the driving circuit based on complex programmable logic devices(CPLDs), and the switching voltage spike of IGBTs can be restrained through software, which is easier and more flexible to adjust. Experimental results demonstrate the effectiveness and practicability of the proposed method.

Frequency-selective Optimization of Periodic Gate Control Signals in DC/DC Converters for EMI-reduction

High-frequency switching of power transistors in power electronic systems can cause electromagnetic emissions.Simple approaches for reducing high-frequency disturbances,such as inserting an additional gate resistor,lead to increased power losses.This makes achieving both electromagnetic compatibility and power efficiency difficult.Active gate drivers help to find a trade-off between these two.Typically,only narrow-band disturbances must be reduced.Accordingly,a target signal with a spectrum notched at some frequencies can be defined.The target signal can be reached by a target-signal-oriented control of the transistor’s gate.This leads to steeper switching slopes,such that the power losses are less increased.Generating arbitrary target signals is impossible.The transistor signal exhibits some physical limitations.A constraint satisfaction problem must be solved,and the gate drive signal must be optimized by applying a residual and Newton’s method.The proposed optimization process in the frequency domain is based on the circuit simulation method named“harmonic balance”.Measurements on a DC/DC converter exhibit the benefits of this method.