The TerraSwarm Research Center

From Murray Wiki
(Redirected from SRC TerraSwarm)
Jump to navigationJump to search

This is a large consortium project led by UC Berkeley as part of the TerraSwarm Research Center (TSRC). This page primarily describes the work done in Richard Murray's group.

Project participants:

  • Sumanth Dathathri (Alumni, CMS)
  • Ioannis Filippidis (Alumni, CDS)
  • Marcella Gomez
  • Scott Livingston*
  • Ivan Papusha
  • Vasu Raman

Collaborators:

The TerraSwarm Research Center (TSRC) aims to enable the simple, reliable, and secure deployment of a multiplicity of advanced distributed sense-control-actuate applications on shared, massively distributed, heterogeneous, and mostly uncoordinated swarm platforms through an open and universal systems architecture.

As new sensing, actuation and data subsystems are added and removed from TerraSwarm Systems, command and control systems must take into the current capabilities to insure that system safety is maintained and performance is optimized. Work at Caltech is focused on the development of architectures and algorithms for specification, design of correct-by-contruction protocols that identify and make use of highly dynamic, networked-based resources to accomplish a task and are capable of providing safe operation based on assume/gaurantee contracts provided by individual subsystems. To manage the complexity of the overall protocol synthesis, we will exploit the structure of assume/gaurantee constructs that provide appropriate separation of concerns and explore the use of incremental techniques that can provide rapid synthesis based on the current system solution. Caltech is also supporting the demonstration the technologies developed on joint demonstration platforms devloped by other members of the TSRC.

Objectives

  • Exploiting distributed localization techniques, we will develop controllers that form and update belief states about the environment without prior detailed models
    • As components move, appear, or disappear, command and control systems must adapt to use currently available services while maintaining system safety and performance.
    • We will develop architectures and algorithms for specification and design of correct-by-construction protocolsthat identify and make use of highly dynamic, networked-based resources to accomplish a task.
    • Safe operation will be ensured using assume/guarantee contracts provided by individual subsystems. A key challenge will be to manage the complexity of protocol synthesis.
  • Define specific applications for use as test cases of TerraSwarm Systems, and perform preliminary analyses of their feasibility. Subsequent years will be devoted to prototyping, deploying, adapting, and testing these applications.

Publications