TARGET DETERMINATION ALGORITHM, PATH PLANNING ALGORITHM, SENSOR SUBSYSTEM, AND COMMUNICATION WITH THE DATABASE IN WHEELED MOBILE ROBOT DESIGN

• Main Author: Arisaputra, Kevin
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/66469
• Institution: Institut Teknologi Bandung

The quarantine process is a common thing to do since the Covid-19 pandemic occurred. In Indonesia, the government provides RSDC (Covid-19 Emergency Hospital) as a place to quarantine Covid-19 patients. In RSDC, patients are generally given consumption 3 times a day. The process of consumption distribution 3 times a day is certainly a repetitive and tiring task to do. In addition, the number of patients who are too many in quarantine can cause disorder during the consumption distribution process. In the case examples that the author found, disorder occurs because consumption is taken in one place. So that other problems can arise during the consumption distribution process, such as the increased risk of disease transmission and the risk of patients not getting consumption. To overcome problems during the consumption distribution process in quarantine, a system is needed that can assist the consumption distribution process. The final project group of authors proposes the use of WMR (Wheeled Mobile Robot) to help carry out the consumption distribution process automatically. The author personally does the work on the target determination algorithm, path planning algorithm, sensors subsystem, and communication with the database. The path planning algorithm itself is implemented by utilizing the BFS (Breadth First Search) algorithm. For the sensors subsystem, it consists of two main sensors i.e., line sensors and obstacle sensors. After all parts have been implemented, there are various conclusions that can be drawn. The line sensor requires a tilt angle of 120? and good tuning for optimal detection results. The obstacle sensor successfully detects an obstacle object with a minimum height of 15.3 cm. The results of line sensor profiling give an average error of ±4.68% and an average standard deviation of 0.31.