SBIR-STTR Award

Current and emerging spaceflight processors are leveraging heterogeneous multicore/co-processor architectures to satisfy the ever increasing onboard processing demands required by planned NASA missions. These architectures can provide increased processing bandwidth, power efficiency, and fault tolerance for onboard processing applications. However, these advantages come at the cost of increased hardware and software complexity. As software development is a major cost driver for missions, this increased complexity has the potential to significantly increase cost for future missions. To address this risk, Troxel Aerospace Industries, Inc. proposes to develop a robust middleware management technology for spacecraft-focused multicore/co-processor architectures. The proposed middleware technology will enable a fault tolerant computing environment that is agnostic to the underlying hardware and is largely transparent to mission applications executing upon the middleware to provide a standardized, intelligent resource, fault, and power management interface.

Potential NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) Troxel Aerospace is committed to continuously pushing the state of the art in spacecraft onboard processing to meet the current and future needs of NASA, the DoD, and commercial customers. The significant improvement in multicore radiation tolerance and co-processor support derived from the successful development of this SBIR effort is of significant interest for all space customers. The Phase I, Phase II, and commercialization plan outlined in this proposal is achievable, has a low risk profile, significantly improves onboard processing capability currently required by multiple NASA missions, and has a clear path to commercialization and acceptance into missions through direct use by programs and infusion via processor vendors who supply to NASA programs. A smooth commercialization path is guaranteed by collaborating with future government and industry customers (i.e. primes and processor vendors) early in Phase I and throughout Phase II to ensure the technology's relevance to their products. The processor-agnostic design of the proposed middleware increases its commercialization potential by being applicable to all processor vendor's present and future technology.

Potential NON-NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) Non-government commercial space customers in all markets will be able to directly benefit from the technology including telecommunications, commercial imagery, launch vehicles, ISS re-service vehicles and exo-planetary commercial ventures (e.g. asteroid mining). Also, non-space markets that require robust fault tolerance from high-performance heterogeneous multicore/co-processor architectures will also be served by the middleware including: radiation test facilities, medical radiation therapy facilities, and other critical performance markets where people's lives or expensive equipment is at risk such as aircraft, automobile, public transportation, and advanced manufacturing where the same processors targeted by the proposed effort are already deployed. Troxel Aerospace plans to capture all these markets by following the a commercialization strategy through partnerships with vendors and primes who already supply processors to these missions and customers.

Technology Taxonomy Mapping:
(NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.) Autonomous Control (see also Control & Monitoring) Computer System Architectures Data Processing Recovery (see also Autonomous Systems) Recovery (see also Vehicle Health Management)

Emerging radiation-hardened and commercial space-capable processors are leveraging general-purpose multicore and niche-application cores to satisfy the ever increasing onboard processing demands required by planned NASA missions. Such architectures can provide increased processing bandwidth and power efficiency for onboard processing applications. However, these advantages come at the cost of increased hardware and software complexity and decreased fault tolerance in the case of commercial technology. As software development is a major cost driver for missions, this increased complexity has the potential to significantly increase cost for future missions. In addition, maintaining mission assurance and fault tolerance is critical. To address these risks, Troxel Aerospace Industries, Inc. (Troxel Aerospace) proposes to continue develop and commercialize a robust middleware management technology for spacecraft heterogeneous multicore processing systems. The middleware technology will enable a fault tolerant computing environment that is portable to different processors and is largely transparent to mission applications executing upon the middleware to provide a standardized, resource-aware, fault tolerant interface for configuration management and heterogeneous resource allocation. This Phase II will include developing the remainder of the middleware, executing a representative application using it across two or three different processor architectures, undertaking a heavy-ion radiation test campaign to quantify its effectiveness in a relevant mission environment, and continuing the commercialization activities begun in Phase I.

Potential NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) The technology under development addresses NASA?s key avionics goals including improved reliability and fault tolerance, increased autonomy, reduced size, weight and power (SWaP), and commonality across spaceflight and ground processing systems put forth in the NASA Crosscutting Technology Roadmap. The technology also enables long-duration crewed missions, space-based observatories, and solar system exploration will require highly reliable, fault-tolerant systems. Communication delays, the challenging orbital dynamics of Near-Earth Asteroids (NEAs), and extreme science missions require increased autonomy for on-board decision infrastructures. Future robotic missions will involve greater complexity and reactivity, which will require increased reliance on autonomy (i.e. advanced onboard processing). Deep-space missions that target active, dynamic, or time-varying phenomena will need robots that can adaptively adjust their configurations and behavior to changing circumstances, and robustly handle uncertainty. Robotic missions to NEAs will require the decision-making and monitoring processes?currently performed by ground control?to be performed by onboard autonomous systems. Advanced avionics technologies and approaches are needed to support these challenging missions and are enabled by Troxel Aerospace?s Robust Multicore Middleware.



Potential NON-NASA Commercial Applications:
:
(Limit 1500 characters, approximately 150 words) All non-NASA government and commercial space customers will directly benefit from the technology by obtaining increased onboard processing capability at reduced budget and SWaP cost for such applications as autonomous operations, improved mission processing, and downlink bandwith management. These features will are enticing to a variety of spacecraft markets including telecommunications, commercial imagery, launch vehicles, ISS re-service vehicles and exo-planetary commercial ventures (e.g. asteroid mining). The small satellite market will be of great interest given their early adoption of modern multicore technology and desperate need to improve system fault tolerance. Also, non-space markets that require robust fault tolerance on high-performance multicore processors will also be served by the middleware including: radiation test facilities (e.g. CERN), medical radiation therapy facilities, and other critical performance markets where people?s lives or expensive equipment is at risk such as aircraft, automobile, public transportation, and advanced manufacturing where the same processors targeted by the proposed effort (e.g. ARM-based processors and GPU co-processors) are already deployed. Troxel Aerospace plans to capture all these markets by following its proven commercialization strategy, i.e. through partnerships with vendors and primes who already supply processors to these missions and customers.

Technology Taxonomy Mapping:
(NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.) Algorithms/Control Software & Systems (see also Autonomous Systems) Autonomous Control (see also Control & Monitoring) Diagnostics/Prognostics Quality/Reliability Recovery (see also Autonomous Systems) Recovery (see also Vehicle Health Management)