Yong-Ho Seo

Mobile Robot Control Using Smart Phone and Its Performance Evaluation

Smart phone that can support mobile computing environment has been popular recently. A new robot application using smart phone is also interesting and feasible research topic at this moment in accordance with a fast growth of mobile internet environment. In this study, we propose a remote control method of a mobile robot using a smart phone with Windows Mobile. We also developed the proposed method using Microsoft’s .Net Compact Framework and Bluetooth data communication to support various OS Platforms from Windows to CE kernel based Embedded OS. In experiment, we evaluated the feasibility and the effectiveness of the proposed method by showing a remote control of a mobile robot using G-Sensor in a smart phone, a line tracing robot application and its performance evaluation.

Remote Data Acquisition and Touch-Based Control of a Mobile Robot Using a Smart Phone

This paper proposes a methodology for remote data acquisition of sensor information and touch-based control of a mobile robot using a smart phone. The acquired sensor information is remotely processed by a smart phone and employed to conduct autonomous navigation. By touching the screen of the smart phone, a series of points obtained from designated curve traces are analyzed and provide control of a robot. This study develops a mobile application that acquires and handles data from a mobile robot and sends appropriate action commands through remote control using Bluetooth communication with a smart phone. The utility and performance of the proposed control scheme have been successfully verified through experimental tasks using an actual smart phone and a mobile robot.

Autonomous Position Estimation of a Mobile Node Based on Landmark and Localization Sensor

This study proposes an efficient position estimation method for localizing a mobile node in indoor environment. Although several conventional methods have been successfully applied for position estimation, they have some drawbacks such as low extendibility in an indoor space, intensive computation, and estimation errors. We propose a precise estimation approach based on a localization sensor and artificial landmarks. In our approach, a mobile node autonomously measures the location of landmarks attached to the ceiling with a localization sensor while moving across the landmarks and building a landmark map. And then, the node estimates its location under the ceiling using the map. In this process, we use a landmark histogram and a Kalman filter to reduce estimation errors. Several experiments performed using a mobile robot successfully demonstrated the feasibility of our proposed approach.

Performance Improvement of Human Detection Using Thermal Imaging Cameras Based on Mahalanobis Distance and Edge Orientation Histogram

In the thermal imaging, human object detection is difficult when the temperatures of surrounding objects are similar to or higher than human’s. In this paper, we propose a novel algorithm suitable to those environments. The proposed method first compute a mean and variance of each pixel value from the initial several frames, assuming that there is no object in those frames. Then for each frame after the initial frames, the Mahalanobis distance is computed between the mean value and the current frame at each pixel, and the region of interest (ROI) is estimated. Finally, using the aspect ratio and edge orientation histogram, we determine if the ROI is human or not. The experimental results show that the proposed method is effective in both summer and autumn.

Remote Monitoring of Patients and Emergency Notification System for U-Healthcare

This study proposes a remote monitoring of patients and emergency notification system with a camera and pulse wave sensor for U-Healthcare. The proposed system is a server client model based U-Healthcare system which consists of wireless clients that have micro-controller, embedded-board for patient status monitoring and a remote management server. The remote management server observes the change of pulse wave data individually in real-time sent from the clients that is to be remote-monitored based on the pulse wave data stored by users and divides them into caution section and emergency section. When the pulse wave data of a user enters an emergency situation, the administrator can make a decision based on the real-time image information and pulse rate variability. When the status of the monitored patient enters the emergency section, the proposed U-healthcare system notifies the administrator and relevant institutions. An experiment was conducted to demonstrate the pulse wave recognition of the proposed system. Key words: Embedded System, Pulse Rate Monitoring, Server-Client Communication, Emergency Detection and Notification

Emergency Detection Based on Motion History Image and AdaBoost for an Intelligent Surveillance System

This paper proposes a method to detect emergency situations in a video stream using a Motion History Image (MHI) and AdaBoost for a video-based intelligent surveillance system. The proposed method creates a MHI of each human object through an image processing technique entailing background removal based on Gaussian Mixture Model (GMM) followed by labeling and accumulating the foreground images. The obtained MHI is then compared with the existing MHI templates to detect emergency situations. To evaluate the proposed emergency detection method, a set of experiments on a dataset of video clips captured from a surveillance camera were conducted. The results show that we successfully detected emergency situations using the proposed method.

Vision-Based Human Following Method for a Mobile Robot Using Human and Face Detection

This paper presents a method of vision based human following for a mobile robot to trace a given human in an indoor environment. For a mobile robot to perform vision based human tracing reliably, we combined a silhouette based human detection and two-stage face detection technique. In this paper, the human detection uses a Histogram of Oriented Gradient (HOG). The face detection is composed of two stages, face detection using a Haar-like wavelet feature and AdaBoost and face detection using skin color with motion information and validation of contour shape. Finally, the performance of the proposed system is verified by conducting experiments where a mobile robot follows a human successfully in a real indoor environment.

Speed Control Using a PID Algorithm for an Educational Mobile Robot

In this paper, we propose a new mothod using the PID controller gain value from rotatation of wheels according to rotation momentum characteristics for an educational mobile robot. P-Gain, I-gain and D-gain have been changed from the mobile robot via wireless remote control in a real-time basis. The transmitted gain value is calculated by real-time operation of embedded software in the robot. Experiments are carried out to find the optimized gain value of two conditions, maximum speed and minimum speed. The gain of intermediate velocity region is calculated from this gain value using primary curve fitting. The intermediate region gain value is calculated by real-time transmission of the mobile robot. Successful results are demonstrated with PID control using the calculated intermediate region gain value.