Informatics Publishing Limited

Muhammad Tayyab ¹, Syed Noman Mahmood ¹, Syed Ashar Alam ¹, Saqib Akhter ¹, Tahir Qadri ¹

Affiliations
1 Sir Syed University of Engineering and Technology

Abstract

This paper proposes and develops a sophisticated design of Autonomous Simultaneous Localization and Mapping (SLAM) based Forklift Robot. The main idea of this work is to design such an object lifting forklift robot which works in dynamic and unknown environment like warehouses to lift and shift the boxes from one place to another. Since the forklift robot is a SLAM base so it has the ability to empower the controller of the robot to take decisions related to the movement of the robot on its own without any human intervention and reach to the objects for detection and lifting. For object detection uses the machine learning approach called Optical Character Recognition (OCR) in which we use K-nearest principle to detect the user define object (box). Our design follows three basic steps: the forward motion planning which is the part of SLAM approach for navigation in forward direction and also generating a real time map. The second step is object detection in which robot match the alphanumeric code written on the box with the code defined by the user to detect and lift that box. The third step is reverse motion planning which follows the map was generated during the forward motion planning to get back its initial position with the box. This map is also display on the user’s interface wirelessly.

Keywords

Forward Motion Planning, Optical Character Recognition, Object Detection, Reverse Motion Planning, Simultaneous Localization and Mapping

Full Text

   

Maqsood Jan Mohammad ¹, Tahir Qadri ¹, Shakil Ahmed ¹

Affiliations
1 Department of Computer Engineering, Sir Syed University of Engineering and Technology, Karachi, PK

Abstract

In Oil and Gas Industry, Multiphase Flow of Oil, Gas and Water is present during exploration, transmission, refining process. Finding the content of oil, gas and water is very important for decision making of exploration, transmission and refining process. Using mechanical instruments is not very efficient, as they are bulky, expensive and not very accurate, so using electronic instruments with embedded system and sensors will capture the image of fluid and then converting that image to digital form. Once the image in digital form, we applied image processing technics of de-blurring to increase the accuracy and then calculating the pixels of oil, gas and water to find content profile. Installing these sensors at two points is used to find velocity profile of fluid. We have successfully fabricated multiphase flow metering (MFM) system using local resources. This system is designed to study the application of wire mesh tomography and for capturing image data of multiphase flow of fluids with the help of embedded system and used de-blurring algorithm of image processing, to increase accuracy of content profile and correlation technique was applied to find velocity profile, then with color image processing techniques, content profile of oil, gas and water was calculated to see that calculation of oil, gas and water are the same and verified as it was obtained from the image processing and de-blurring techniques. Results obtained were quite encouraging.

Keywords

Correlation, De-blurring, Embedded System, Flow Control, Image Processing, Multiphase

Full Text