Chinthaka Premachandra

Road crack detection using color variance distribution and discriminant analysis for approaching smooth vehicle movement on non-smooth roads

We present an image analysis based automatic road crack detection method for conducting smooth driving on non-smooth road surfaces. In the new proposal, first the road surface areas which include cracks are extracted as crack images by analyzing the color variance on norm. Then cracks are extracted from those areas by introducing a method on discriminant analysis. According to experiments using the images of different road surfaces, the new proposal showed better performances than the conventional approaches.

Outdoor Road-to-Vehicle Visible Light Communication Using On-Vehicle High-Speed Camera

We propose a road-to-vehicle Visible Light Communication (VLC) system for ITS. Here, the communication between road and a vehicle is approached using an LED array as a transmitter and an on-vehicle high-speed camera as a receiver. Modulation is conducted at the transmitter by blinking LEDs in high frequency and demodulation is conducted at the receiver by capturing the blinking patterns of the transmitter, by processing the high-speed camera images captured in high frame rate. This paper proposes an effective image processing method for capturing those blinking patterns. Out door communication experiments of the proposed VLC system are also present.

Development of an air-ground operational robot and its fundamental controlling approach

A lot of studies can be found in the literature for ground operational robot such as humanoid robot, wheel robot, snake type robot and so on. On the other hand, some studies can be found in the literature for indoor and outdoor air operational robot development such as small type helicopter and multi copter controlling. If a robot can conduct both air and ground operations, it is easy to conduct effective operations in the fields like natural disaster areas. Thus, a robot which enables to achieve air-ground operations is developed. This paper presents development of main mechanical structure of the robot. Furthermore, a fundamental autonomous moving and flying control step of the robot is also described.

High performance embedding environment for reacting suddenly appeared road obstacles

We present a high performance embedding environment for developing vision based vehicle/ robot controlling. This particular environment is consisted of two main parts as recognition part and control part. Former can conducts vision based recognition targets such as obstacle detection while latter can controls the vehicle/robot in real time following the recognition information. The tests were conducted to confirm the performance of the proposed embedding environment, installing it on a wheel robot. Here, the test was conduct to stop the moving robot when an obstacle appeared suddenly. In the experiments, system could conduct obstacle detection as well as motor controlling effectively in order to stop the robot.

Speed-up Ellipse Enclosing Character Detection Approach for Large-size Document Images by Parallel Scanning and Hough Transform

This paper presents a speed-up ellipse enclosing character detection algorithm that uses parallel image scanning and the Hough transform (HT) for large-size document images. Objects in images are generally detected based on geometrical information obtained via raster scanning. In raster scanning, all pixels of an image are scanned starting from the upper-left point and ending with the lower-right point. In the case of large-size images, considerable time is needed for processing an image by scanning all pixels. In this paper, an object detection approach for large-size images is proposed which does not require scanning all pixels in the images. In this speed-up ellipse enclosing character detection approach for large-size document images, pixels are scanned on constantly spaced vertical parallel lines. If an object larger than a certain size is identified while scanning, the presence of an ellipse enclosing character is assumed and ellipse detection is conducted by applying HT only in a defined local image area over the found object. With this approach, processing time can be dramatically reduced by disregarding some objects and reducing the total image area used for ellipse detection.

Speedy Character Line Detection Algorithm Using Image Block-Based Histogram Analysis

Academic institutions such as universities and technical colleges usually employ paper-based examinations and reports to evaluate the academic performance of students. Consequently, teachers expend considerable time and energy in the marking of such paper-based examinations. We are developing an automatic paper marking system geared towards reducing this paper-marking burden on teachers. To execute paper marking, handwritten character lines are extracted from examination papers, and then characters on those lines are recognized. In this paper, we primarily discuss how the character line is extracted from handwritten examination papers without ruled lines. The extraction of character lines from non-ruled papers is difficult because of the writing characteristic of students. Further, extraction accuracy is an important factor in character recognition performance. Conventional character line extraction algorithms for printed documents perform poorly on this problem. Furthermore, most proposed methods conduct tests using document images that include only character lines. In this paper we develop a less time-consuming algorithm for this task.

New controller design method based on the data conversion method

For targeting a feedback system with an unknown plant (controlled object), there are several methods for controller designing. In order to reduce costs, conventional methods use a single set of experimental I/O data to design a controller. In this paper, we propose a new method for this purpose, which improves the conventional data conversion method (DCM) and enables us to design a controller by giving an outline figure of the ideal output. In the proposed method, two kinds of the I/O data of the plant are defined. One data is obtained from the linearity of the plant, experimental data (single set of I/O), and an unknown vector. The other is obtained from the structure of a feedback system when the ideal output is given. A controller is designed by matching these two data. Because designing with giving the ideal output directly is possible, this method provides a design environment that makes it possible to reflect a users idea. Furthermore, proposed method produces a more appropriate parameter for the controller. We confirmed the effectiveness of new method by conducting simulation experiments. In the experiments, we compared proposed method with conventional DCM, and designed a controller by assuming the outline of the ideal output. The result of proposed method showed better performances compared to the conventional methods.

Smooth automatic vehicle stopping control system for unexpected obstacles

Various studies have been conducted regarding vehicles and obstacle avoidance, but very few studies deal with the avoidance of obstacles that suddenly and unexpectedly appear during vehicle operation. We conduct automatic stopping control of the vehicle using a fuzzy control system when the vehicle stop suddenly. The propose system conducts stopping control of the vehicle depending on the distance to the obstacle. The simulation experiments and actual experiments using a mobile robot on behalf of a vehicle were conducted regarding the proposal.

Suppressive Fair Buffer Management Policy for Intermittently Connected Mobile Ad Hoc Networks

We propose a suppressive fair buffer management policy for intermittently connected mobile ad-hoc networks. So far, several buffer management policies have been considered. These existing buffer management policies assume that all stored messages can be replaced when nodes encounter each other. Buffer management policies, however, can prioritize messages stored at the receiving side over those at the sending side. By doing this, message transmissions are suppressive, and thus energy consumption in terms of sending messages is reduced. Moreover, our proposed policy gives relay messages with a small number of message copies to high priority. Specifically, our proposed policy maintains a sharing of buffer spaces that is as fair as possible. In this paper, we reveal how the suppression of receiving messages affects the system performance compared with existing buffer management policies. Through simulation experiments, we show that the suppressive fair buffer management policy improves energy consumption without largely degrading the delivery failure probability and the mean delivery delay.

Adaptive navigation and motion planning for a mobile track robot

Localization and navigation of mobile robots precisely in an indoor environment is one of most important and challenging tasks. Navigation using conventional approaches such as Global Positioning System (GPS) and some vision based odometry are not very much effective in indoor environments. Therefore, a magnetic wheel encoding mechanism was selected in order to improve the navigational method. Also it could be with the most common localization approaches such as GPS, inertial navigation systems and laser sensors. This paper discusses a mobile robot application which navigates using magnetic wheel encoders and camera sensor. Further, the robot uses Wi-Fi to gather information and it creates intelligence to find the path dynamically. The Central control center and its main database process all available data and send the relevant control commands to the robots. Additionally, it will direct the messages between the nodes and the robot and keep robots on the correct track. The environment that the robot is moving is a pulse of black lines in white background. For the detection of the line Hough transformation has been used. It is having the capability of detecting sensor nodes using Radio-frequency identification (RFID) or Quick Response (QR) codes and taking measurements and returning to the base station avoiding obstacles communicating between nodes and the base.