RISC: Current Research

Distributed Autonomy

Details

Collaborative missions

The goal is to develop algorithms to enable a group of mobile autonomous systems to achieve a collaborative task using only onboard sensing, computation, and communication, i.e., there is no external coordinator. Example missions include area coverage and mutiagent pursuit-evasion.

 

  • M. Abdelkader, U.A. Fiaz, N. Toumi, M.A. Mabrok, and J.S. Shamma, "Cooperative multi-agent system for autonomous aerial grasping in outdoor environments". video

  • M. Abdelkader, Y. Lu, H. Jaleel, and J.S. Shamma, "Distributed real time control of multiple UAVs in adversarial environment: Algorithm and flight testing results", International Conference on Robotics and Automation, Brisbane, Australia, May 2017. (to appear)


Self-organizing systems

In self-organization, the objective is to design protocols for a potentially large number individual components to evoke desirable collective outcomes. Our work includes self-assembly, under which large numbers of individual components form and break bonds to build desirable structures, and area coverage, where mobile sensors self-deploy to cover an unknown region. We also are interested in theory to support algorithmic performance guarantees..

             
  • M.J. Fox and J.S. Shamma, "Probabilistic performance guarantees for distributed self-assembly", IEEE Transactions on Automatic Control, December 2015, pp. 3180–​3194. link​
  • A.Y. Yazicioglu, M. Egerstedt, and J.S. Shamma, "Communication-free distributed coverage for networked systems", IEEE Transactions on Control of Network Systems, September 2017, pp. 499510. link
  • H. Jaleel and J.S. Shamma, "Transient response analysis of metropolis learning in games", 20th World Congress of the International Federation of Automatic Control, July 2017. link
  • A. Yasin Yazicioglu, M. Egerstedt, and J.S. Shamma,"Formation of robust multi-agent networks through self-organization of random regular graphs", IEEE Transactions on Network Science and Engineering, October-December 2015, pp. 139–151. link​

Distributed optimization

In distributed optimization, a group of decision making nodes seeks to compute the outcome of an optimization problem. The data for the optimization problem is distributed over the collection of nodes, meaning that nodes must communicate to be able to compute the optimization. We are interested in algorithms that exploit special structures for real-time implementability. Applications include multi-robot motion coordination.

 
  • H. Jaleel and J.S. Shamma, "Distributed submodular minimization and motion coordination over discrete state space"​.
  • H. Jaleel and J.S. Shamma, "Decentralized energy aware co-optimization of mobility and communication in multiagent systems", 55th IEEE Conference on Decision and Control, December 2016. link
  • H. Jaleel and J.S. Shamma, "Design of real-time implementable distributed suboptimal control: An LQR Perspective", IEEE Transactions on Control of Network Systems", accepted 2017. (to appear) link