Publication

Exploration in Inaccessible Terrain Using Visual and Proprioceptive Data

Alexander Dettmann, Daniel Kuehn, Dominik Van Opdenbosch, Tobias Stark, Heiner Peters, Sankaranarayanan Natarajan, Sebastian Kasperski, Arne Böckmann, Adrian Garcea, Eckehard Steinbach, Frank Kirchner

In: Proceedings of the 14th International Symposium on Artificial Intelligence, Robotics and Automation in Space. International Symposium on Artificial Intelligence, Robotics and Automation in Space (iSAIRAS-2018) 14th June 4-6 Madrid Spain 6/2018.

Abstract

Mountains, gorges and caves on celestial bodies are particularly intriguing for scientists. In order to explore such a demanding environment, a swarm of heterogeneous robots seems to be promising. In this paper, a heterogeneous robotic team consisting of a quadruped and a rover is presented. For cost-effective exploration, a mission control is implemented which plans robot-specific trajectories on a common map. Although the robotic team members are described briefly, the focus lies on navigation algorithms based on proprioceptive and visual data provided by the robots internal sensors as well as on a panoramic and a time of flight camera, respectively. The results show good mapping and localization capabilities, thus being a cost-effective alternative to traditional stereo camera-based navigation.

20180424_Exploration_in_Inaccessible_Terrain_Using_Visual_and_Proprioceptive_Data.pdf (pdf, 572 KB)

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz