AgreeOnIT: Lättvikts konsensus och distribuerat datakunskap i resursbegränsade sakernas Internet
Forskningsprojekt , 2019 – 2022

The Internet of Things (IoT) and its novel applications pose unique challenges: many of its applications are mission- or safety-critical, have stringent deadlines, and require complex coordination. In cooperative driving, autonomous vehicles have to agree wirelessly on their planned trajectories within milliseconds. Similarly, a modern factory has thousands of sensors, actuators and control loops which need to coordinate within tight deadlines. Missing such deadlines can have substantial consequences including system failure and threats to human life.  Such applications demand dependable algorithms for consensus at a message latency and energy efficiency that today´s approaches cannot provide. The challenge roots in the mismatch between the complex nature of consensus algorithms and the contrasting resource-constraints and dynamics of low-power wireless links of IoT Systems.This project, AgreeOnIT, will enable a paradigm shift from wired to wireless coordination in mission- and safety-critical applications in IoT. Based on my recent advances on synchronous transmissions and distributed computing in resource-constrained environments, I will take an interdisciplinary approach and devise: (1) new methods for distributed computing in resource-constrained environments; and (2) lightweight consensus algorithms enabling network-wide agreements at ultra-low-latency in dynamic networks. AgreeOnIT will improve the current state of the art by more than two orders of magnitude.

Deltagare

Olaf Landsiedel (kontakt)

Gästforskare vid Chalmers, Data- och informationsteknik, Nätverk och system

Christos Profentzas

Doktorand vid Chalmers, Data- och informationsteknik, Nätverk och system

Finansiering

Vetenskapsrådet (VR)

Finansierar Chalmers deltagande under 2019–2022

Relaterade styrkeområden och infrastruktur

Informations- och kommunikationsteknik

Styrkeområden

Grundläggande vetenskaper

Fundament

Publikationer

Mer information

Senast uppdaterat

2019-10-07