An Approach for Autonomous Multi-rover Collaboration for Mars Cave Exploration: Preliminary Results. Vaquero, T., Troesch, M., & Chien, S. In International Symposium on Artificial Intelligence, Robotics, and Automation in Space (i-SAIRAS 2018), Madrid, Spain, June, 2018. Also appears at the 28th International Conference on Automated Planning and Scheduling (ICAPS) 2018 Workshop on Planning and Robotics (PlanRob) and Workshop on Distributed and Multi-Agent Planning (DMAP), Delft, Netherlands.
abstract   bibtex   
Mars caves are promising targets for planetary science and human shelter. Exploring these environments would pose several challenges, including limited communication, lack of sunlight, limited vehicles lifetime that would not allow humans in the loop, and a totally unknown environment. Mission to these underground environments would required levels of autonomy, coordination and collaboration never been deployed before in rovers. In this paper we propose a multi-rover coordination algorithm and experimental framework for cave exploration missions. We describe preliminary experimental results with this coordination algorithm in a realistic simulated cave. We analyze rover coordination performance in different environmental settings and provide insights on potential opportunities for enhanced autonomy with AI planning and scheduling.
@inproceedings{vaquero-troesch-chien-Cave-i-SAIRAS-2018,
	title        = {An Approach for Autonomous Multi-rover Collaboration for Mars Cave Exploration: Preliminary Results},
	author       = {Tiago Vaquero and Martina Troesch and Steve Chien},
	year         = 2018,
	month        = {June},
	booktitle    = {International Symposium on Artificial Intelligence, Robotics, and Automation in Space (i-SAIRAS 2018)},
	address      = {Madrid, Spain},
	note         = {Also appears at the 28th International Conference on Automated Planning and Scheduling (ICAPS) 2018 Workshop on Planning and Robotics (PlanRob) and Workshop on Distributed and Multi-Agent Planning (DMAP), Delft, Netherlands.},
	abstract     = {Mars caves are promising targets for planetary science and human shelter. Exploring these environments would pose several challenges, including limited communication, lack of sunlight, limited vehicles lifetime that would not allow humans in the loop, and a totally unknown environment. Mission to these underground environments would required levels of autonomy, coordination and collaboration never been deployed before in rovers. In this paper we propose a multi-rover coordination algorithm and experimental framework for cave exploration missions. We describe preliminary experimental results with this coordination algorithm in a realistic simulated cave. We analyze rover coordination performance in different environmental settings and provide insights on potential opportunities for enhanced autonomy with AI planning and scheduling.},
	clearance    = {CL\#18-2123},
	project      = {CaveRovers},
	todo         = {url}
}

Downloads: 0