This paper presents algorithms to navigate and avoid obstacles for an in-doorautonomous mobile robot. A laser range finder is used to obtain 3D images ofthe environment. A new algorithm, namely 3D-to-2D image pressure and barriersdetection (IPaBD), is proposed to create a 2D global map from the 3D images.This map is basic to design the trajectory. A tracking controller is developedto control the robot to follow the trajectory. The obstacle avoidance isaddressed with the use of sonar sensors. An improved vector field histogram(Improved-VFH) algorithm is presented with improvements to overcome somelimitations of the original VFH. Experiments have been conducted and the resultis encouraged.
展开▼
