Adaptive intelligent systems for energy management of vehicles

The thesis demonstrated the architecture of adaptive intelligent systems for energy management that is capable of interacting with complex systems including the vehicle, environment, and driver components, as well as the interrelationships between these variables, to deliver fuel consumption improvements.

Intelligent clothing for automated recognition of human physical activities in free-living environment

This paper presents an intelligent clothing framework for human daily activity recognition using a single waist-worn tri-axial accelerometer sensor coupled with a robust pattern recognition system. The activity recognition algorithm is realized to distinguish six different physical activities through three major steps: acceleration signal collection/pre-processing, wavelet-based principle component analysis, and a support vector machine classifier. The proposed activity recognition method has been experimentally validated through two batches of trials with an overall mean classification accuracy of 95.25 and 94.87%, respectively. These results suggest that the intelligent clothing is not only able to learn the activity patterns but also capable of generalizing new data from both known and unknown subjects. This enables the proposed intelligent clothing to be applied in a comfortable and in situ assessment of human physical activities, which would open up new market segments to the textile industry.

Look-ahead intelligent energy management of a parallel hybrid electric vehicle

Improving fuel efficiency in vehicles can reduce the energy consumption concerns associated with operating the vehicles. This paper presents a model for a parallel hybrid electric vehicle. In the model, the flow of energy starts from wheels and spreads toward engine and electric motor. A fuzzy logic based control strategy is implemented for the vehicle. The controller manages the energy flow from the engine and the electric motor, controlling transmission ratio, adjusting speed, and sustaining battery's state of charge. The controller examines the vehicle speed, demand torque, slope difference, state of charge of battery, and engine and electric motor rotation speeds. It then determines the best values for continuous variable transmission ratio, speed, and torque. A slope window method is formed that takes into account the look-ahead slope information, and determines the best vehicle speed. The developed model and control strategy are simulated using real highway data relating to Nowra-Bateman Bay in Australia, and SAE Highway Fuel Economy Driving Schedule. The simulation results are presented and discussed. It is shown that the use of the proposed fuzzy controller reduces the fuel consumption of the vehicle.

Intelligent 3D programmable surface

Creating a highly programmable surface operating at relatively high speed and in real time is an area of research with many challenges. Such a system has applications in the field of optical telescopes, product manufacturing, and giant 3D-screens and billboards for advertising and artwork. This paper covers certain aspects of a keynote presentation at ISDT 2010 including system design, modularity, programmability and the system control intelligence. An overview of the system architecture, actuator design, electronics and distributed control will provide an insight into how the system is controlled and self-tuned for a number of applications. A simulation environment that has been developed to streamline system reconfiguration will also be presented, demonstrating translation of complex mathematical functions into 3D shapes virtually before being displayed on the physical surface.

Optimal algorithm for a mobile intelligent gas detection device

This thesis is based on the development of a gas detection device that can be mounted on a mobile robotic platform. The focus was on development of the A.I recognition algorithm with an array of sensors to detect trace amounts of explosive and volatile gases in the environments it is exposed to.

Intelligent energy management for plug-in hybrid electric vehicles : the role of ITS infrastructure in vehicle electrification

The desire to reduce carbon emissions due to transportation sources has led over the past decade to the development of new propulsion technologies, focused on vehicle electrification (including hybrid, plug-in hybrid and battery electric vehicles). These propulsion technologies, along with advances in telecommunication and computing power, have the potential of making passenger and commercial vehicles more energy efficient and environment friendly. In particular, energy management algorithms are an integral part of plug-in vehicles and are very important for achieving the performance benefits. The optimal performance of energy management algorithms depends strongly on the ability to forecast energy demand from the vehicle. Information available about environment (temperature, humidity, wind, road grade, etc.) and traffic (traffic density, traffic lights, etc.), is very important in operating a vehicle at optimal efficiency. This article outlines some current technologies that can help achieving this optimum efficiency goal. In addition to information available from telematic and geographical information systems, knowledge of projected vehicle charging demand on the power grid is necessary to build an intelligent energy management controller for future plug-in hybrid and electric vehicles. The impact of charging millions of vehicles from the power grid could be significant, in the form of increased loading of power plants, transmission and distribution lines, emissions and economics (information are given and discussed for the US case). Therefore, this effect should be considered in an intelligent way by controlling/scheduling the charging through a communication based distributed control.

A hybrid intelligent system and its application to fault detection and diagnosis

This paper proposes a hybrid system that integrates the SOM (Self Organizing Map) neural network, the kMER (kernel-based Maximum Entropy learning Rule) algorithm and the Probabilistic Neural Network (PNN) for data visualization and classification. The rationales of this hybrid SOM-kMER-PNN model are explained, and the applicability of the proposed model is demonstrated using two benchmark data sets and a real-world application to fault detection and diagnosis. The outcomes show that the hybrid system is able to achieve comparable classification rates when compared to those from a number of existing classifiers and, at the same time, to produce meaningful visualization of the data sets.