Search

In this paper we propose a novel system that uses sensory input from both vision and inertial sensors for an improved perception of the robot current status of equilibrium. We use MonoSLAM vision odometry as a basis for the visual perception and a gyro for angular velocity measurements; and we devise a reweighting method within a Kalman filter framework. Moreover, our approach is designed to be robust against visual and measurement noise such as blur, poor lighting conditions, and faulty sensory output. The novelty in this work is a robust humanoid fall avoidance system, which relies on the fusion of sensory input, mainly gyroscope and visual odometry, taking into account changes in the environment. The fusion of the mentioned sensors in addition to the image quality assessment, ensure a more human-like fall avoidance in comparison to currently existing systems. We implement our method on the NAO humanoid, where seven sets of experiments are performed to assess the effectiveness of our approach. The fusion of camera and gyro information not only enables a more human-like behavior, but also provides more humanoid stability and faster recovery, and thus leads to more robust fall avoidance. [ABSTRACT FROM AUTHOR]

Mobile applications are becoming computationally intensive nowadays due to the increasing convenience, reliance on, and sophistication of smartphones. Nevertheless, battery lifetime remains a major obstacle that prohibits the large-scale adoption of such apps. Mobile cloud computing is a promising solution whereby apps are partially processed in the cloud to minimize the overall energy consumption of smartphones. However, this will not necessarily save energy if there is no systematic mechanism to evaluate the effect of offloading an app onto the cloud. In this paper, we present a mathematical model that represents this energy consumption optimization problem. We propose an algorithm to dynamically solve the problem while taking security measures into account. We also propose the free sequence protocol (FSP) that allows for the dynamic execution of apps according to their call graph. Our experimental setup consists of an Android smartphone and a Java server in the cloud. The results demonstrate that our approach saves battery lifetime and enhances performance. They also show the effects of workload amount, network type, computation cost, security operations, signal strength, and call graph structure on the optimized overall energy consumption. [ABSTRACT FROM AUTHOR]

Abstract: Support vector machines (SVM) have been showing high accuracy of prediction in many applications. However, as any statistical learning algorithm, SVM's accuracy drops if some of the training points are contaminated by an unknown source of noise. The choice of clean training points is critical to avoid the overfitting problem which occurs generally when the model is excessively complex, which is reflected by a high accuracy over the training set and a low accuracy over the testing set (unseen points). In this paper we present a new multi-level SVM architecture that splits the training set into points that are labeled as ‘easily classifiable’ which do not cause an increase in the model complexity and ‘non-easily classifiable’ which are responsible for increasing the complexity. This method is used to create an SVM architecture that yields on average a higher accuracy than a traditional soft margin SVM trained with the same training set. The architecture is tested on the well known US postal handwritten digit recognition problem, the Wisconsin breast cancer dataset and on the agitation detection dataset. The results show an increase in the overall accuracy for the three datasets. Throughout this paper the word confidence is used to denote the confidence over the decision as commonly used in the literature. [Copyright &y& Elsevier]

Abstract: Forest fire occurrence prediction plays a major role in resource allocation, mitigation and recovery efforts. This paper compares two artificial intelligence based methods, artificial neural networks (ANN) and support vector machines (SVM), utilizing a reduced set of weather parameters. Using a reduced set of parameters results in an efficient and reduced cost prediction system especially for developing countries. In this paper the aim is to predict forest fire occurrence by reducing the number of monitored features, and eliminating the need for weather prediction mechanisms. The reason is to reduce errors due to inaccuracies in weather prediction. The challenge is to choose a limited number of easily measurable features in the aim of reducing the cost of the system and its maintenance. At the same time, the chosen features must have a high correlation with the risk of fire occurrence. A literature review of forest fire prediction methods divided into systems/indices, and artificial intelligence is provided. The two fire danger prediction algorithms utilize relative humidity and cumulative precipitation to output a risk estimate. The assessment of these algorithms, using data from Lebanon, demonstrated their ability to accurately predict the risk of fire occurrence on a scale of four levels. [Copyright &y& Elsevier]