Autonomous Navigation and Obstacle Avoidance For Service Robot

Abstract

The dramatic increase in the elderly population, along with the explosion of costs, poses problems for the community. Therefore, an alternative way is needed to provide care to the elderly. Robot technology at the same time, it is undergoing progress triggered by increased computing and a decrease in the cost of sensor technology. In the application of robot service for elderly people, we need a robot that is able to follow the elderly people automatically, so the robot's operation would be disrupted if the environment changed, for example, if there are new obstacles on the robot track or the robot is moved to another room. The problem faced by an autonomous mobile robot is mapping, localization and navigation. These three elements are SLAM building elements. This study aims to overcome these problems. In this research, the mapping and localization process was carried out using the library RTABMap. For navigation, the path that the robot must pass from a start point to goal point is created using the A* (A-star) algorithm. Whereas to avoid obstacles, we use the Dynamic-Window Approach and costmap algorithms. Based on the results of the study, the robot has succeeded in mapping in a new environment that the robot did not know before, localizing it using visual features of the environment, detecting new obstacles that were not previously on the map, applying the A* algorithm to do path planning and automatically navigate using the path plan that has been created.