The widespread adoption of cutting-edge 5G/6G wireless networks together with advancements in learning technologies is giving rise to a flourishing population of connected autonomous systems that now inhabit our daily lives. Internet of Things, swarms of Unmanned Aerial Vehicles, fleets of autonomous driving vehicles. As the inexorable march of technology plunges us forward, our coexistence with massive numbers of connected autonomous agents seems imminent, bound to change our world. It is cities, magnets of human population, that bear the brunt of this paradigm shift. Unmanned aerial vehicles crowding the skies and self-driving cars dominating the streets. These marvels of connected autonomy promise all forms of new services, but at what price?

As human-operated unconnected technology becomes autonomous and ubiquitously connected, a new panorama of market and societal possibilities opens before us. However, harnessing these opportunities goes beyond a mere pursuit of maximizing utilities. Connected autonomous systems must operate with a holistic consideration for the human element that is now excluded from their operation. While the greedy pursuits of a solitary agent can be curtailed, as we transition to large numbers of connected agents, such restraints become elusive. It is imperative to develop distributed and connected algorithms that can impose requirements and constraints to their behavior. We must move beyond local single-objective learning problems and embrace a more holistic approach, where distributed requirements can be explicitly specified and guarantees of their satisfaction can be obtained. That is, to make connected autonomous systems behave in a manner that respects the desires and requirements of society, stakeholders and policy makers

LOCALS (Learning Opportunities for Connected Autonomy in Livable Spaces) attempts to grapple with some of the intricate technical challenges that lie at the heart of this problem. Proposing algorithms, providing guarantees and numerically and experimentally validating the resulting behavior of the connected autonomous agents.