SLAM Based Indoor Autonomous Navigation System For Electric Wheelchair

Abstract

At present manual and electric wheelchairs are widely used in the market. Manual wheelchairs rely on other people's or their own power to drive wheelchairs, which consumes physical strength, while electric wheelchairs will cause driving fatigue and limited cognitive ability. This paper introduces an indoor electric wheelchair autonomous navigation system. The wheel encoder and IMU (Inertial Measurement Unit)are used for wheelchair pose estimation, which avoids the large error caused by using a single sensor for pose estimation, At the same time, the dynamic obstacles are scanned by Lidar, and the map is constructed by combining the surrounding walls and gmapping algorithm. Finally, the autonomous navigation based on move_base framework is realized. The system uses STM (STMicroelectronics) 32 series chip as the core of the bottom hardware processor, Raspberry pie as the upper processor, and embedded ROS (Robot Operating System) to easily and safely integrate with the existing electric wheelchair. Can well solve the elderly reaction, cognitive ability is limited by its defects such as their own and others' safety problem, and the electric wheelchair has a lot of the human-computer interaction interface, such as face, voice, gestures and so on, so as to enrich the daily life of the elderly, and implement more than one modal for subsequent intelligent wheelchair nursing platform provide strong technical support.