AI supported Multi-Functional Gripping System for Dexterous Manipulation Tasks

Abstract

Dexterous mobile manipulators that are capable of performing a wide array of tasks are essential for unstructured human-centered environments. Especially in rescue scenarios where time and resources are limited, a gripping system should be as versatile but at the same time as efficient as possible. In addition, situational awareness in accidents and impaired visibility is essential for safely performing missions or automating behaviors, and thus supporting operators. The main contribution of this work is a multi-functional gripping system with technologies and methods for manipulation in rescue and recovery operations as well as for handling hazardous materials. The gripping system is supplemented by an RGB and thermal camera and object detection algorithms, which run in real-time on an embedded device for robust recognition in harsh and dynamic environments. The proposed multi-functional gripping system has been thoroughly evaluated and tested in laboratory experiments and real field facilities.