The purpose of this project is to organize a two-day workshop on multi-robot systems that addresses the practical aspects and issues involved in real-world implementation of these systems. The field of multi-robot systems has seen a tremendous analytical progress that has included development of mathematical and algorithmic tools to enable information processing, control, and communication. These systems have tremendous potential for applications in dynamic and uncertain conditions that characterize a wide range of civilian and military applications such as disaster management, large-scale monitoring and surveillance, and space exploration. Before these systems can be operational in real-world environments, not only a number of research issues have to be resolved but also the theories generated from research have to be verified, validated, and tested by extensive experimentation and field implementation. The workshop focuses on five issues central to the experimental and real-world implementation aspects of multi-agent robotics that includes: i) multi-robot hardware; ii) software integration; iii) communication; iv) sensing; and v) simulation environments. The main objectives of the workshop include determining strategic research directions, developing action plans, and foment interdisciplinary collaborations to help overcome the shortcomings prevalent in the current systems.
The workshop brings together researchers working on the multi-disciplinary aspects of the swarm robotics such as systems and controls, communication, sensor data processing and fusion, and development of sensors. In addition to researchers working in the engineering disciplines, researchers from several other disciplines such as biology, physics, and mathematics will be invited to discuss various aspects of this research. Such a forum enables the cross-fertilization of ideas from multiple scientific domains and promotes interdisciplinary research. This project enables and supports participation of students, especially those at undergraduate levels and from underrepresented groups. Elaborate arrangements are made to widely disseminate the presentations, discussions, and findings from this workshop to broader research community.
Workshop Theme: A number of future civilian and military missions will rely on networked groups of vehicles and sensors operating in dynamic and uncertain conditions with little or no direct supervision. Unmanned Aerial Vehicles (UAVs), Unattended Ground Sensors (UGSs), and Unmanned Ground Vehicles (UGVs) working together possess crucial qualities such as performance, adaptability, and robustness to uncertainties and failures that far exceed the achievable qualities of individual systems. These networked systems have generated a great deal of interest in a wide range of application domains such as disaster monitoring and management, large-scale surveillance, detection and tracking, search and rescue, and perimeter protection. Before these systems can be operational in real-world environments, not only do a number of research issues have to be resolved, but the theories generated from research have to be verified, validated, and tested by extensive experimentation and field implementation as well. The Pratt School of Engineering at Duke University hosted the workshop entitled "Frontiers of Real-World Multi-Robot Systems: Challenges and Opportunities" on October 10-11, 2011. The workshop engaged researchers working in this area with the following objectives: i) to evaluate the current state-of-the-art of experimental platforms available at various university and federal laboratories; ii) to discuss current and future real-world civilian and military applications enabled by multi-robot systems; iii) to identify the shortcomings of these platforms that prevent implementation of multi-robot systems; and iv) to determine strategic research directions, develop action plans, and foment interdisciplinary collaborations to help overcome these shortcomings. Intellectual Merit: The workshop brought together experts from a number of educational institutions and federal laboratories involved in various aspects of theoretical and applied research in the area of multi-robot systems. Twelve leading researchers spoke on topics including experimental platforms, theoretical aspects, communication aspects, and software and testing issues. Furthermore, breakout sessions were held at the end of the workshop where the researchers discussed the pressing issues in the area of multi-robot systems. The discussions held at the workshop after the presentations and during the break-out sessions highlighted the need for research in a number of directions including algorithmic developments; design of robust, reliable, and resilient systems; science of testing; human-robot systems; and novel hardware designs. Broader Impacts: The workshop attracted 53 participants from academia, industry, and government, including 20 students who were encouraged to present posters at the workshop. Funds were used to partially support travel of 17 of these students. A website (www.min.uc.edu/mrw ) has been created for wide dissemination where the presentation slides are uploaded so that interested researchers can have access to these documents. Outcomes: A final report was submitted to the NSF that detailed the technical recommendations based on the discussions held at the workshop. The report recommended future research in four areas: i) Science of Testing; ii) Perception and Control; iii) Human-aided Co-ordination; and iv) Communication. The workshop was well received with all participants surveyed rating the overall quality as very good or excellent.