Heterogeneity, Diversity and Resilience
in Multi-Robot Systems

August 15-16, 2016
Westin Arlington, VA


The timeliness of this workshop is granted by our aspirations to solve increasingly complex problems that require a distribution of distinct capabilities among robot team members. During the process of developing such solutions, heterogeneity and diversity become design features. However, we still lack the necessary analytical tools that help us understand the implications. By building on foundational concepts of network science and security, we should aim at developing formal definitions of robustness, resilience and privacy that can be applied to networked multi-robot systems. The workshop’s scope of topics will explore the various dimensions along which heterogeneity and diversity occur: behavioral, functional, morphological, temporal, and spatial.

Background

The last five years have seen advances in flexible and rapid fabrication methods for monolithic, multi-material, and multi-scale manufacturing, which are enabling low cost, customized robots. The cost of communication and storage technologies are falling at 20-30% every year. Sensors, actuators, and processors are being integrated into a global information technology network, enabling a diverse array of robots and accompanying software and hardware infrastructure. However, such large teams of heterogeneous robots cannot be efficiently programmed individually, and yet, humans must be able to supervise, guide and coach such systems. More importantly we lack the understanding to design heterogeneous multi-robot systems and the tools to analyze the resiliency of such systems. Indeed, our understanding of diversity is quite limited.

Objectives

This workshop will bring together experts who will address the key issues of designing heterogeneous robot systems, analyzing resiliency in multi-robot networks, and modeling diversity in robot-robot or human-robot teams. By building on foundational concepts of network science and security, the objective is to develop formal definitions of robustness, resilience and privacy that can be applied to networked multi-robot systems. We believe that the resulting research will significantly impact the field of heterogeneous multi-robot systems. Participants will be expected to contribute papers that will be published in a book, which will be the first monograph to address the above-mentioned ideas.

Topics of Interest

  • Quantification of diversity, heterogeneity measures
  • Analysis of robustness and resilience in multi-robot systems
  • Privacy and security of data in large-scale robot networks
  • Control / coordination strategies for heterogeneous robot teams
  • Perception and sensor fusion in heterogeneous robot teams
  • Specialization, task allocation and role assignment
  • Communication and symbolic languages in heterogeneous teams
  • Morphological heterogeneity in modular robotic systems
  • Accidental heterogeneity caused by drift (SW) or wear-and-tear (HW)

Organizers

Dr. Amanda Prorok
University of Pennsylvania
 
Dr. Brian M. Sadler
Intelligent Systems Army Research Laboratory
 
Prof. Vijay Kumar
University of Pennsylvania