The goal of this project is to create an integrative framework for the design of coupled biological and robotic systems that accommodates system uncertainties and competing objectives in a rigorous, holistic, and effective manner. The design principles are developed using a concrete, end-to-end application of tracking and modeling fish movement with a network of gliding robotic fish. The proposed robotic platform is an energy-efficient underwater gliding robotic fish that travels by changing its buoyancy and mass distribution (gliding) or by flapping tail fin (swimming).
After Berkeley, Berlin, Seattle, Vienna, and Pittsburgh, we are this time in beautiful Porto, Portugal, again co-located with the annual Cyber-Physical Systems Week.
After Berkeley, Berlin, Seattle, Vienna, and Pittsburgh, we are this time in beautiful Porto, Portugal, again co-located with the annual Cyber-Physical Systems Week.
Full paper submission deadline is 4th February 2018. Accepted and presented papers will be submitted to IEEE Xplore digital library.
As multi-agent systems become ubiquitous, the ability to satisfy multiple system-level constraints in these systems grows increasingly important. In applications ranging from automated cruise control to safety in robot swarms, barrier functions have emerged as a tool to provably meet such constraints by guaranteeing forward invariance of a set. However, satisfying multiple constraints typically implies formulating multiple barrier functions, bringing up the need to address the degree to which multiple barrier functions may be composed through Boolean logic.
This paper studies the multi-agent average consensus problem under the requirement of differential privacy of the agents' initial states against an adversary that has access to all the messages. We first establish that a differentially private consensus algorithm cannot guarantee convergence of the agents' states to the exact average in distribution, which in turn implies the same impossibility for other stronger notions of convergence.
Airborne networking utilizes direct flight-to-to-flight communication for flexible information sharing, safe maneuvering, and coordination of time-critical missions. It is challenging because of the high mobility, stringent safety requirements, and uncertain airspace environment. This project uses a co-design approach that exploits the mutual benefits of networking and decentralized mobility control in an uncertain heterogeneous environment.
