Hi Diogenese,
Victoria da Poian is a NASA Visiting Scientist, whose actual employer is a company called Microtel LLC, specialising in software and systems engineering applications for Earth science and space missions, that was acquired by Tyto Athene in April.
https://science.gsfc.nasa.gov/sci/bio/victoria.dapoian
View attachment 69367
View attachment 69368
Tyto Athene Acquires Microtel to Expand Space System Capabilities; Dennis Kelly Quoted - GovCon Wire
Tyto Athene Acquires Microtel to Expand Space System Capabilitieswww.govconwire.com
The new ML algorithm you referred to was developed specifically for data collected, analysed and sent back to Earth (!) by an instrument aboard the Rosalind Franklin rover, the Mars Organic Molecule Analyzer (MOMA) - said algorithm will be used by researchers on Earth to study the data and decide about the next steps, not by the instrument autonomously.
And it has nothing to do with Rosalind Franklin’s autonomous navigation system at all.
View attachment 69369
NASA Trains Machine Learning Algorithm for Mars Sample Analysis - NASA
When a robotic rover lands on another world, scientists have a limited amount of time to collect data from the troves of explorable material, because of shortwww.nasa.gov
View attachment 69370
Besides, there is no indication that Victoria Da Poian has been working with SNNs and/or TENNs in recent years.
You should also ask yourself the following: If she had been blown away by Akida technology or TENNs, why would she and her colleagues have set up two open-science ML challenges in 2022 and published a paper about the results in 2023 (published in early 2024), declaring CNN solutions the winning entries in both cases?
Leveraging open science machine learning challenges for data constrained planetary mission instruments
Abstract. We set up two open-science machine learning (ML) challenges focusing on building models to automatically analyse mass spectrometry (MS) data foracademic.oup.com
View attachment 69377
Furthermore, here is a link to the agreement signed by ESA and NASA in May:
https://www.nasa.gov/news-release/nasa-european-space-agency-unite-to-land-europes-rover-on-mars/
“With this memorandum of understanding, the NASA Launch Services Program will procure a U.S. commercial launch provider for the Rosalind Franklin rover. The agency will also provide heater units and elements of the propulsion system needed to land on Mars.”
All available evidence points to NASA having nothing to do with the ESA Mars rover’s navigation system.
This would highly likely be in the sole hands of Airbus Defence & Space that designed and built the rover now known as the ExoMars Rosalind Franklin rover in a cleanroom at its UK facility in Stevenage, before it was transferred to Italy in preparation for its planned launch in September 2022 from Baikonur, the spaceport operated by Russia within Kazakhstan.
While the Airbus team from Stavanage recently tested two of their rovers in a nearby quarry, including one similar to Rosalind Franklin nicknamed Charlie that - according to the following ESA blog article - was used “to test a new navigation system that could speed up rover missions on the Moon and Mars. This capability could lead to improved performance for ESA’s ExoMars Rosalind Franklin’s rover mission, which is expected to launch in 2028.”
https://blogs.esa.int/to-mars-and-back/2024/09/06/rovers-lightsabres-and-a-piglet/
View attachment 69378
… I doubt we are involved here.
Listen to this June 2023 presentation by Airbus Space Robotics Tech Lead Piotr Weclewski from 20:50 min, when he mentions the effort of validating new approaches such as neuromorphic or bio-inspired computing (that are generally of great interest to Airbus): “We need to bring it slowly, we need to bring it step by step to the picture. Why? Because of the criticality of the missions."
He goes on to emphasise that sending something not only to low-Earth orbit, but to the Moon, dramatically increases the complexity and the risk of radiation mitigation, thermal aspects etc.
So disruptive approaches such as the implementation of neuromorphic technology need to be carefully validated first in order not to endanger critical missions. That definitely sounds to me as if a trip to the Moon will be a first step only after successful validation of this new technology in Low-Earth orbit (LEO), and that a mission to Mars, which is even more complex, will remain a long-term goal for now.
Here is another article on the quarry testing conducted earlier this month that quotes Airbus rover program manager Chris Draper, confirming that the Moon is the initial target for their rovers being tested these days:
View attachment 69380
The step-by-step validation approach ties in with the new ESA R&T program referred to as NEURAVIS, which - as we found out in July - will be conducted by a consortium consisting of Airbus, BrainChip, Frontgrade Gaisler and Neurobus.
This must have been the relevant tender that was awarded to them, titled
EVALUATION OF NEUROMORPHIC COMPUTING TECHNOLOGIES FOR VERY LOW POWER AI/ML APPLICATIONS - EXPRO PLUS
https://esastar-publication-ext.sso.esa.int/ESATenderActions/details/60869+
View attachment 69375
View attachment 69374
All this information combined with last week’s LinkedIn comments by both Alf Kuchenbuch and Laurent Hili (that I had quoted in my previous post, but which you didn’t address at all) strongly suggest to me we are not/will not yet be involved in the Rosalind Franklin Rover Mars mission. And IMO that includes TENNs software as a separate offering. Why else would Alf Kuchenbuch have referred to his “dream” of Laurent Hili, Sandi Habinc and Sean Hehir collaborating on providing Rosalind’s “children” with “a neuromorphic brain for better perception and autonomy”? (“That is my dream.”)
As the BrainChip staff member primarily dealing with our company’s space industry partners, he’d know exactly whether or not TENNs were currently being tested by Airbus in order to improve Rosalind Franklin’s navigation system. If that were indeed the case, why would he then refer to a new generation of rovers only and not to the Rosalind Franklin rover itself?
TENNs will first need to be validated, too.
So to sum it up, I predict there will be validation of our technology in LEO satellites first, before ESA will consider implementing it on the more critical and complex Lunar or even Martian missions.
Episode 6 Quarterly Sean said we have engagements on all of AKIDA1000, 1500 and GEN 2+TENNs. ESA awarded Airbus Neuravis and AKIDA 1500......
I guess we know who one of the engagements is with.
