We have long been looking for official confirmation of a successful NASA Vorago Brainchip Phase 1 project.
Well I have found it and it is copied with a link below.
Note:
1. Vorago is Silicon Space Technology Corporation for newer shareholders.
2. Vorago used the term CNN RNN to describe AKIDA not SCNN.
As you read through the extracts you will note the following:
A. Vorago met all of the Phase 1 objectives
B. Vorago has five letters in support of continuing to the next Phase 2 importantly/interestingly two of these letters offer funding for the Phase 2 independent of NASA - (I personally am thinking large Aerospace companies jumping on board)
C. Vorago has modelling which shows AKIDA will allow NASA to have autonomous Rovers that will achieve speeds of up to 20 kph compared with a present speed of 4 centimetres a second.
There is in my opinion no other company on this planet with technology that can compete with the creation of Peter van der Made and Anil Mankar.
The Original Phase 1:
"The ultimate goal of this project is to create a radiation-hardened Neural Network suitable for Ede use. Neural Networks operating at the Edge will need to perform Continuous Learning and Few-shot/One-shot Learning with very low energy requirements, as will NN operation. Spiking Neural Networks (SNNs) provide the architectural framework to enable Edge operation and Continuous Learning. SNNs are event-driven and represent events as a spike or a train of spikes. Because of the sparsity of their data representation, the amount of processing Neural Networks need to do for the same stimulus can be significantly less than conventional Convolutional Neural Networks (CNNs), much like a human brain. To function in Space and in other extreme Edge environments, Neural Networks, including SNNs, must be made rad-hard.Brainchip’s Akida Event Domain Neural Processor (
www.brainchipinc.com) offers native support for SNNs. Brainchip has been able to drive power consumption down to about 3 pJ per synaptic operation in their 28nm Si implementation. The Akida Development Environment (ADE) uses industry-standard development tools Tensorflow and Keras to allow easy simulation of its IP.Phase I is the first step towards creating radiation-hardened Edge AI capability. We plan to use the Akida Neural Processor architecture and, in Phase I, will: Understand the operation of Brainchip’s IP Understand 28nm instantiation of that IP (Akida) Evaluate radiation vulnerability of different parts of the IP through the Akida Development Environment Define architecture of target IC Define how HARDSIL® will be used to harden each chosen IP block Choose a target CMOS node (likely 28nm) and create a plan to design and fabricate the IC in that node, including defining the HARDSIL® process modules for this baseline process Define the radiation testing plan to establish the radiation robustness of the ICSuccessfully accomplishing these objectives:Establishes the feasibility of creating a useful, radiation-hardened product IC with embedded NPU and already-existing supporting software ecosystem to allow rapid adoption and productive use within NASA and the Space community.\n\n\n\n\t Creates the basis for an executable Phase II proposal and path towards fabrication of the processor."
CNN RNN Processor
FIRM: SILICON SPACE TECHNOLOGY CORPORATION PI: Jim Carlquist Proposal #:H6.22-4509
NON-PROPRIETARY DATA
Objectives:
The goal of this project is the creation of a radiation-hardened Spiking Neural Network (SNN) SoC based on the BrainChip Akida Neuron Fabric IP. Akida is a member of a small set of existing SNN architectures structured to more closely emulate computation in a human brain. The rationale for using a Spiking Neural Network (SNN) for Edge AI Computing is because of its efficiencies. The neurmorphic approach used in the Akida architecture takes fewer MACs per operation since it creates and uses sparsity of both weights and activation by its event-based model. In addition, Akida reduces memory consumption by quantizing and compressing network parameters. This also helps to reduce power consumption and die size while maintaining performance.
Spiking Neural Network Block Diagram
ACCOMPLISHMENTS
Notable Deliverables Provided:
• Design and Manufacturing Plans
• Radiation Testing Plan (included in Final report)
• Technical final report
Key Milestones Met
• Understand Akida Architecture
• Understand 28nm Implementation
• Evaluate Radiation Vulnerability of the IP Through the Akida
Development Environment
• Define Architecture of Target IC
• Define how HARDSIL® will be used in Target IC
• Create Design and Manufacturing Plans
• Define the Radiation Testing Plan to Establish the Radiation
Robustness of the IC
FUTURE PLANNED DEVELOPMENTS
Planned Post-Phase II Partners
We received five Letters of Support for this project.
Two of which will provide capital infusion to keep the project going, one for aid in radiation testing, and the final two for use in future space flights.
Planned/Possible Mission Infusion
NASA is keen to increase the performance of its autonomous rovers to allow for greater speeds.
Current routing methodologies limit speeds to 4cm/sec while NASA has a goal to be able to have autonomous rovers traverse at speeds up to 20km/hr.
Early calculations show the potential for this device to process several of the required neural network algorithms fast enough to meet this goal.
Planned/Possible Mission Commercialization
A detailed plan is included in the Phase I final submittal to commercialized a RADHARD flight ready QML, SNN SoC to be available for NASA and commercial use.
This plan will include a Phase II plus extensions to reach the commercialization goals we are seeking.
CONTRACT (CENTER): SUBTOPIC:
SOLICITATION-PHASE: TA:
80NSSC20C0365 (ARC)
H6.22 Deep Neural Net and Neuromorphic Processors for In- Space Autonomy and Cognition
SBIR 2020-I4.5.0 Autonomy
My opinion only DYOR
FF
AKIDA BALLISTA
www.qualcomm.com
