ROS-Based Indoor Autonomous Exploration and Navigation Wheelchair

Aiming at the current situation of high cost, complicated construction and poor reusability of the autonomous navigation system, an indoor wheelchair autonomous exploration and navigation system with low cost and high reusability is realized in this paper. The host computer program is based on the Robot Operating System (ROS), an extensible Arduino microcontroller is used to control wheelchair motors. The wheelchair adopts two-wheel differential control structure. Compared to expensive laser sensors, a cost-effective RGB-D camera is used to perceive environment. The depth point cloud from RGB-D camera is cast to laser data and then used for map building and obstacle detection of Gmapping algorithm. The position and orientation of wheelchair are estimated by Adaptive Monte Carlo Localization (AMCL) algorithm. A* algorithm is applied to global path planning; Dynamic Window Approach (DWA) algorithm is used for local path planning. Frontier-Based approach for autonomous exploration is used to explore unknown regions automatically. Finally, an Android APP is developed to provide friendly human-computer interaction interfaces.