Line tracking control of royal Thai air force nursing mobile robot using visual feedback

Abstract

The specific objective of this study was to design a Royal Thai Air Force (RTAF) nursing robot, so-called Thadloom, to assist medical personnel in delivering food, medical supplies and communicating remotely with patients during the COVID-19 pandemic. This study also seeks to develop many identical mobile robots to be used in concerned places, such as hospitals or clinics, to reduce the risk of infection to medical personnel when entering the controlled area COVID-19-patients. This paper presents the design and implementation of real-time vision systems for the RTAF nursing robots when used in the controlled area of COVID-19-patients, with a capacity of delivering food and medical supplies automatically. The robots were able to follow lines on the floor using visual feedback. A contour tracing algorithm was implemented for line identification and position estimation. Linear segments with parabolic blends were used for the robot’s trajectory planning. The robot can track a line on the floor effectively by employing a PID controller. This experimental result showed that the developed vision-based state estimation and the designed PID controller. It was also found that it received a good satisfaction towards the accuracy of the position within 3.5 centimeters, i.e., The distance from the center of the robot to the center of the line.