Seminari su Sistemi multi-robot
A Symbiotic Human and Multi-Robot Planetary Exploration System
Speaker: Dr. Jacopo Panerati
The availability of next generation heavy launcherssuch as NASAs SLS and SpaceXs Falcon Heavywill enable new planetary exploration missions. Most space agencies are now targeting the Moon as the next step in exploration beyond LEO and many already have plans for precursor robotic and human exploration. For example, ESA has championed the Moon Village concept since 2016 and NASAs 2018 budget includes a Lunar Exploration Campaign. In this context, natural caves are appealing solutions to shelter humans and equipment for long-duration Lunar missions. In 2017, data from JAXAs Kaguya probe revealed a 50km-long lava tube. For safety reasons, preliminary robotic exploration is imperative. Multi-robot systems carry the potential for greater efficiency and higher fault-tolerance, because of their ability to cooperate and inherent redundancy. Including humans into these systems is desirable to mitigate the complexity of the system but challenging at the interface level. The presentation, in particular, will focus on the preliminary results obtained during ESA's recent PANGAEA-X planetary analog campaign in Lanzarote (Spain). We designed experiments to address two main subject: (i) the distributed computation of the multi-robot network properties; and (ii) the human-multi-robot system interaction.
How do you program 1000 robots?
Speaker: Prof. Giovanni Beltrame
We are currently on the verge of a new technology revolution---autonomous robots are becoming more and more present in our everyday lives. From drones to self-driving cars, these systems are becoming pervasive, and are acting as an enabling technology for many kinds of safety-critical applications. Examples of robotic applications are search-and-rescue operations, industrial and agricultural inspection, autonomous car driving, aerial mapping, monument digitization, and surgery. Despite this ambitious vision, the major achievements in the area of swarm robotics still consist of algorithms that tackle specific problem instances, and the performance of these algorithms strongly depends upon the context in which they are developed. Given this state of affairs, reproducing results and comparing algorithms is difficult, hindering the development of swarm robotics as a whole. We present a novel paradigm for the development of complex swarm behaviors. It offers a small, but powerful set of operations to specify behaviors both in a swarm-wide fashion, and from the point of view of an individual robot. This swarm-oriented programming offers the promise of letting a designer program thousands of robots in a manageable way. This talk will present the overall swarm-oriented approach, as well as practical examples in the area of disaster response.