Cloud-Based Efficient Data Analysis and Reduction (CEDAR)

Period of Performance: 06/25/2015 - 09/19/2017

$750K

Phase 2 SBIR

Recipient Firm

Architecture Technology Corp.
9971 Valley View Road Array
Eden Prairie, MN 55344
Firm POC
Principal Investigator

Abstract

ABSTRACT:CEDAR is a software solution implementing a cloud computing capability within an unmanned aircraft system (UAS) such as the MQ-9 Reaper. It aggregates the available processing and storage capacity of onboard computers enabling them to be shared across a variety of mission applications ranging from sensor exploitation (SE) applications to on-board diagnostics and prognostics applications where the applications may belong to different security domains, e.g., Unclassified, and SECRET. By enabling mission-specific smarts to be easily programmed within a UAS, CEDAR provides the foundation for implementing the needed autonomy for mission operations. CEDAR allows a mix of computer platforms with varying capabilities to be dynamically provisioned among independently developed software applications in isolated security domains. It does this while providing fast, direct communication of sensor data, usually without the need for cross-domain guards, and without compromising security. The CEDAR design supports mission commanders in planning software deployment pre-mission to ensure available cloud hardware supports mission software needs. It enables autonomous in-mission cloud provisioning decisions by cloud-located applications, allowing the cloud to react to hardware failures and mission-generated tasks without requiring ground control intervention, making the CEDAR cloud appropriate for permissive and A2AD/contested environments. BENEFIT:CEDAR is a novel software product that efficiently and autonomously manages shared compute resources in an unmanned entity (UAV or robot) to run mission applications that analyze, combine, and filter data in real-time or near real-time. CEDAR provides an open platform for supporting the development of new UAS and Internet of Things applications. By supporting deployment of simultaneous applications in multiple security domains, it maximizes compute and storage resource utilization, allowing diverse independent, single-focus applications and providing the performance necessary for UAS on-board sensor exploitation with the possibility of increased UAS autonomy. Autonomous cloud control ensures mission-critical applications are provided critical resources, and allows testing new applications in an active platform without risk of interfering with existing mission-critical applications. The result of CEDAR research and development will be software adaptable to diverse hardware platforms, suitable for managing a cloud of low-cost COTS computer devices hosting commercial real-time operating systems and providing mixed capabilities in terms of processing power and coprocessor support that will benefit a wide variety of sensor data analysis algorithms. CEDAR will facilitate rapid innovation of surveillance, sensor exploitation, and diagnostics/prognostics applications by presenting a consistent virtualized environment from development to operational use. CEDARs cloud provisioning approach supports autonomous operation of UAV and robotic platforms, decreasing the oversight required Potential customers for the resulting commercial product include the military, law enforcement and homeland defense, and commercial enterprises.