Diversity and Resilience in Robot Networks. Diversity is a powerful concept. Yet, we have not yet shown how diversity affects the performance of engineered systems, such as robot networks. Moreover, as we build heterogeneous multi-robot systems with interdependent entities, we must take care to ensure resilience. This talk presents some recent results.
Solving the traffic problem. In this contribution by Bloomberg, we show some of the work we have been doing on cooperative formation control for increasing traffic efficiency and safety (featuring Michael Whitzer, Monroe Kennedy, and myself).
Control of a heterogeneous robot swarm. How can we efficiently program a large group of heterogeneous robots to redistribute according to specialized task requirements? Our control solution implicitly solves the combinatorial problem of distributing the right number of robots of a given ‘species’ to the right tasks. A key characteristic of our policy is its scalability with respect to the number of robots, the number of species, and the number of traits.
Diversity in engineered systems. In this interview, we describe our work on heterogeneous robot swarms. Why is ‘diversity’ so important? Can we measure it, and correlate it with the resulting performance of the system? The movie was made to the occasion of our Best Paper Award at BICT 2015.
An internet of robots and things. This video demonstrates how a robot running ROS is can access the Internet of Things via ‘accessors’, which are software modules developed in the Ptolemy project at UC Berkeley. This was the first demo to demonstrate a functioning interface between ROS and accessors.
Multi-robot localization. This (artsy) video demonstrates the collaborative localization capabilities of a group of mobile robots using noisy relative range an bearing sensors developed at the DISAL Lab, EPFL. The position estimate is computed on board the robots and sent to an overhead projector for display.