Robust Mapping and Path Planning for Indoor Robots Based on Sensor Integration of Sonar and a 2D Laser Range Finder

Abstract

In this study a sensor integration scheme to improve the capabilities of an indoor robot, has been developed such as mapping, collision avoidance and path planning. This integration has been carried out using a sonar ranging system and a 2D laser scanner, with participation of other sensors. The main goal of the integration is to reinforce the robustness of the overall system and overcome the sensors' disadvantages. This system inherits the advantages of these sensors and enables the robot to navigate safely in the working environment. The 2D laser scanner has been used to develop a mapping technique called the vector mapping algorithm to get consistent high precision maps with low memory requirements. The generated maps have been used as a base for an autonomous path planning algorithm depending on straight line navigation (SLN algorithm). The vector mapping algorithm has been implemented on an RWI B21 robot, one of the available robot platforms at the University of Tbingen.