Kristóf Csorba

Analyzing computation offloading energy-efficiency measurements

No one disputes that mobile phones have become part of everyday life. Besides phone calling we use them for browsing, messaging, playing games and many other things which were possible only on a desktop computer several years ago. Although these devices are even more “smart”, but due to the growing resource requirements their batteries are discharging in a very short period. This is a very important phenomenon, which forces to find alternative ways to reduce the energy consumption of the smartphones. One of them is the method of “computation offloading” where a part of the processes are executed on a remote device (e.g. in the cloud). The creation of an energy-efficiency model consists of several steps. One of the hardest part is to understand how a smartphone behaves in different circumstances. During the execution of an application every step has its own energy cost. To make the execution energy-efficiency, we need first to measure and analyse these costs. In this paper we are going to present a measurement system which is used to analyse the energy consumption of smartphones. The measurements were made in different scenarios and the goal was to save energy through offloading some tasks. The offloading process is based on scheduling theory.

Target Tracking and Surrounding with Swarm Robots

Nowadays the capabilities of autonomous robots have increased rapidly. However there are situations where one robot is not sufficient. In this paper we will introduce a completely distributed algorithm for tracing and surrounding moving objects or targets using homogenous, medium scaled swarm robots based on local sensing. Our new approach is based on the already known concept of basis behaviors as ubiquitous general building blocks for synthesizing artificial group behavior. Our goal was to generate an algorithm based on simple rules in order to protect a territory from a human intruder. In contrast with the common solution which is employing potential fields, we tried to keep our method as simple as possible by utilizing rule-based control mechanism. We will also present our virtual simulation environment and experimental results.

Energy-efficient computation offloading model for mobile phone environment

The popularity of smartphones is growing every day. Thanks to the more powerful hardware the applications can run more tasks and use broadband network connection, however there are several known issues. For example, under typical usage (messaging, browsing, and gaming) a smartphone can be discharged in one day. This makes the battery life one of the biggest problems of the mobile devices. That is a good motivation to find energy-efficient solutions. One of the possible methods is the “computation offloading” mechanism, which means that some of the tasks are uploaded to the cloud. In this paper we are going to present a new energy-efficient job scheduling model and a measurement infrastructure which is used to analyze the energy consumption of smartphones. Our results are going to be demonstrated through some scenarios where the goal is to save energy. The offloading task is based on LP and scheduling problems.

The RobonAUT Autonomous Mobile Robot Construction Contest

Abstract This paper presents the RobonAUT contest organized by the Budapest University of Technology and Economics, Department of Automation and Applied Informatics. The contest aims to support the education of control theory, embedded system design and several other courses related to building autonomous mobile robots in a fun and instructive teamwork project. Students are required to build autonomous four-wheel robots capable to go along a path indicated by a continuous line on the ground, and to perform several tasks along the way. These tasks involve obstacle avoidance, parking operations, passing slopes, and acceleration segments. 3 students work together in a team during the semester to build their robot, for which they need skills related to control theory, both analog and digital circuit design, and microcontroller programming. The task has a very strong integrating nature by utilizing the material of several courses of electrical engineering. Beside the motivating fact that building an own robot is fun, we believe that students are given the opportunity to apply their knowledge to solve a real-world engineering task. And based on the popularity of the challenge, students seem to realize the need for this practice as well. This paper is aimed to describe the history, rules and experiences of the challenges in 2010 and 2011, together with the plans of the 2012 challenge.

Vision Based Area Discovery with Swarm Robots

Nowadays the need for autonomous robots has increased rapidly. These robots should have different types of sensors, thus their functionality can be extended. In order to process the data coming from the sensors, the robots require a proper processing hardware. However, the cost of a unique on-board controller system, which is able to handle advanced algorithms, is relatively expensive. Our proposal is to attach an inexpensive commercial mobile device to the robot, which has a high-resolution camera and advanced processing unit for different types of operations. In this paper we will present a vision-based model for environment discovery using homogeneous, medium scaled swarm of robots by using the on-board camera of the controller-smart phone. We introduce an extended axiom system based on bio-inspired swarm models, which optimizes the energy consumption of the swarm.

Improved Topic Identification for Similar Document Search on Mobile Devices

This paper presents a novel, two level classifier ensemble designed to support document topic identification in mobile device environments. The proposed system aims at supporting mobile device users who search for documents located in other mobile devices which have similar topic to the documents on the users own device. Conforming to the environment of mobile devices, the algorithms are designed for slower processor, smaller memory capacity and they maintain small data traffic between the devices in order to keep low the cost of communication. We propose a keyword list based topic comparison, enhanced with a two level classifier ensemble to accelerate the topic identification process. The new technique enables document topic comparison using few communication traffic and it requires few calculations.

Marker localization with a multi-camera system

This paper presents a marker localization system consisting of multiple cameras. It is the basis of a real-time robotic motion tracking system aiming for high localization accuracy and high time resolution. These goals are achieved using several cameras with very redundant field-of-views. The paper presents the theoretical background of 3D camera calibration and localization, the localization accuracy measurement setup, and its results. The measurements are obtained using a color marker and an industrial robotic arm for measurement ground truth generation. The Smart Mobile Eyes for Localization (SMEyeL) project is open-source: the source code, all measurement input data and documentation are public available.

Educational aspects of designing robot for Eurobot contest

Abstract The Eurobot contest is an international robot building competition. This paper presents an autonomous robot mechanical and electric design built to Eurobot 2011 robotics contest. The robot collects the playing elements, builds towers and avoid collision with the opponent robot. The building process, the control and vision algorithms are also presented. We specify the used absolute (ultrasonic sonar) and relative (odometry) positioning methods, and we describe their advantages and disadvantages. The hardware and software architecture, navigation algorithms are also presented. This paper focuses on the educational aspects of the project. We demonstrate how our team collaborates, and how the tasks were assigned to them in order to comply their thesis.

Edge Preserving Range Image Smoothing by Rotated Bilateral Sampling

In this paper we deal with (step and crease) edge preserving smoothing and normal estimation in low depth-resolution range images. As a result of short base distance stereo, compression or scene geometry, such range images consist of few discrete range values. We propose a method for smoothing such range images by interpolation while preserving boundary and crease edges, and estimating surface normals. The proposed method is based on rotated weighting sampling matrices and iso-range curves extracted from the range data. Samples and weights are selected using a rotated weight matrix and planar surfaces are fitted. Based on the fitting error, the best fit is selected to estimate the surface normal and interpolate surface points in such sparse dataset. Fitting error distribution is also used for estimating crease edge score. Such low-level understanding of range images can be utilized in further processing steps such as segmentation, localization, mapping, object detection etc. The output of the algorithm is presented and evaluated using both simulation and real range image data. It is shown that such sampling preserves crease edges and provides lower estimation errors for interpolated range image points.

Optimizing camera placement for localization accuracy

This paper presents the optimization of a camera placement for improved localization accuracy. It is the basis of a localization system aiming for high localization accuracy. This goal is achieved using several cameras with very redundant field-of-views. The paper presents the calculation of the localization accuracy - which depends on the camera model and the pixel quantizing error - at one specific point. A method is introduced for handling areas in a probabilistic sense instead of examining only one point. The accuracy can be improved by adding a new camera to the system. The calculation of the optimal position of the new camera at compliance to some restrictions is demonstrated. The Smart Mobile Eyes for Localization (SMEyeL) project is open-source: the source code, all measurement input data and documentation are public available.