Swarm-based Subterranean Exploration

点击翻译按钮获取中文摘要

In this program we will develop robot swarms that collaborate to implement rapid mapping, localization, and navigation in limited-access and communication-restricted areas. To maximize the flexibility of our approach, we will consider a heterogeneous robotic team composed of both unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs). The distinctive trait of our approach is the use of a dynamic, ad-hoc communication network, collaborative localization, and fully decentralized decision making and mapping.******We will focus on: ***1) Communication: to allow the robots to communicate as they explore and***map the environment, we will use gossip communication. Maps, tasks, and commands will be diffused using a robust approach similar to that of distributed databases, but requiring low bandwidth. A connectivity maintenance controller will be embedded in the robots, so that they can execute explore the environment without getting separated from the team and risk being lost;***2) Localization: we propose a decentralized collaborative localization approach based on range-only Ultra-Wide Band (UWB) sensors and vision that does not require any prior infrastructure. We will use communication to achieve a consensus on a global coordinate frame, UWB ranging for relative localization of the robots, and visual odometry for absolute localization in the environment;***3) Mapping: the mapping of the area will be achieved using a collaborative vision- and laser-based techniques, merging the sensed data into a single map. This map will then serve to locate hot spots, conduct searches, and locate the swarm with respect to points of interest. Our initial results in collaboration with the Masachussetts Institute of Technology are very promising.******Our research output will be tested in collaboration with the European Space Agency (ESA) and, possibly, the Canadian Space Agency (CSA) in the test site of Tiguaton, on the island of Lanzarote, Spain. We request four types of robots for performing our research on exploration and mapping of GPS-denied environments. As our research is axed on the collaboration between ground and aerial robots, we request both rovers and quadcopters. Recently, our team had a high degree of success in developing many swarm-robotics algorithms, many of which were tested on datasets or in laboratory conditions only because we don't have access to the equipment necessary for field deployment. Nowadays it is already a challenging to attract excellent students, therefore we believe we must provide them with equipment that allows to fully develop their research projects to have a substantial impact on the research landscape. Considering 8 students close to final year, time is of essence: advancements in the field may be lost without the requested equipment. In addition, students would be deprived of a unique learning opportunity in Tinguaton, and Canada would lose its only participant in the campaign.**