BRN Discussion Ongoing

[Frangipani]

Hi @Bravo,

is that by any chance your Canadian holiday home?

May 30, 2023 - Ottawa, Ontario

Valcartier test range, Interconnect Bravo-Bravo basecamp. Credit: Michel Guitard, science visual documentation at DRDC.

 

[Andy38]

Drone detection prototypes involving neuromorphic event-based cameras are already being tested - a perfect future use case for Akida. The Canadian researchers in the article below used a DVXplorer event camera, after having previously experimented with a DAVIS 346 event camera, both made by Swiss company iniVation - see https://dl.acm.org/doi/pdf/10.1145/3546790.3546800 - published Sept 7, 2022. While we can practically exclude that Akida was used in the prototypes described, the Canadian researchers are concluding: “Moreover, we will continue to follow the improvement of neuromorphic hardware.” (taken from the just quoted PDF)

So in case of a possible collaboration between iniVation and Brainchip (that I had wondered about in a previous post, after noticing promotion of their new Aeveon sensor technology was not mentioning their former partner SynSense, while at the same time using images of a wallaby - of all animals - as illustration), a future event-based iniVation camera might well contain Akida.

Then of course there is Prophesee as another manufacturer of neuromorphic cameras that is already partnering with Brainchip. And lots of armed forces worldwide interested in this technology. I would be very surprised if Akida would ultimately not be taken into consideration for this type of drone detection.

Eye on the sky: new drone detection technology advances national security - National Research Council Canada

nrc.canada.ca

Eye on the sky: new drone detection technology advances national security​

May 30, 2023 - Ottawa, Ontario

Valcartier test range, Interconnect Bravo-Bravo basecamp. Credit: Michel Guitard, science visual documentation at DRDC.

Drones take up a lot of airspace around the world these days—filming movie scenes, delivering goods, gathering agricultural data, supporting search-and-rescue operations as well as conducting military surveillance, targeting and attack. Their sizes can range from small recreational units that fit into the palm of your hand to military drones weighing upwards of 600 kilograms. And the commercial drone market is expected to grow from over US$20 billion today to US$500 billion by 2028.

As their numbers surge, uncrewed aerial systems (UASs), commonly known as drones, will pose more hazards than ever, whether planned or unplanned. When photographing weddings and events, they could encroach on the public's privacy. While flying over airports, prisons or military facilities, they could compromise security. And in war zones, they can pose danger to lives, homes and infrastructure.

"Over the past year, we've seen a rapid evolution of UAS use on the battlefield in Ukraine," says Andrew Scheidl, Program Lead of the Multimedia Analytics Tools for Security program at the National Research Council of Canada (NRC). "Those developments will affect future deployments of the Canadian Armed Forces, but they will also migrate to other threat actors. Reliable detection and countermeasure systems will be increasingly important for military and public safety applications."

Autel Evo II FLIR drone. Credit: Michel Guitard, science visual documentation at DRDC.

In other high-profile incidents, the world has seen a drone crash onto the White House lawn, several circle around a nuclear power plant in France and others bomb a Ukrainian army weapons warehouse.

With the number of scenarios for illegal drone activities growing every day, the need for innovative drone-detection systems is intensifying. And in combat zones, having the ability to identify and counter enemy drones is particularly important.

A longstanding collaboration between the NRC and Defence Research and Development Canada (DRDC) sparked the development of a new approach to drone detection that disrupted the status quo, one that uses AI and classifies drones by their propeller speeds. The highly skilled team has brought all the necessary expertise to the table: optics, physics, signal processing, machine learning and vision, and neuromorphic systems.

AI helps dodge the perils and pitfalls of drone detection​

Matrice 200 drone carrying NRC prototype visual frequency detection system that was tested at Interconnect Bravo-Bravo basecamp. Credit: Michel Guitard, science visual documentation at DRDC.

While several methods of detecting drones have been in use for a long time, none are totally dependable, particularly in dense urban areas and forests. This is because radio frequency, acoustic and optical detection systems can be misled by noise in complex environments and cause false or missed detections. For example, tall buildings and trees can with interfere with the ability of visible-light and infrared cameras to match the appearance of a live drone to images in a large database of UAS models. As well, while ground-based radars detect drones efficiently, their reliability can be affected by the environment and geography. Using these radar systems is also an expensive approach that requires bulky equipment and a lot of power. In addition, some of them transmit an active signal, thus exposing the devices.

Close-up Autel Evo II drone controller. Credit: Michel Guitard, science visual documentation at DRDC.

Over the past 4 years, an R&D team of engineers and scientists from the NRC's Digital Technologies Research Centre and the DRDC have developed an innovative technological solution for passively spotting drones in cluttered settings. It incorporates AI to accurately detect, track and characterize drones on the basis of the signal generated by their rotating propellers rather than by using an image bank. This method generates very few false alarms and accurately detects low-flying drones that use topography to evade discovery. The "signature" of a drone propeller can also be used to discriminate or classify aircraft by type, such as identifying it as a helicopter and not a drone.
The DRDC team initially lab-tested the feasibility of passive detection using drone characteristics. The next step was to develop energy-efficient detection algorithms and predict the performance of systems using different hardware. After some ground testing, the team built the first prototypes by combining the hardware and software and replacing some of the physical modelling of sensor responses with machine learning and AI technologies.

Changing the drone-detection game​

For a week in October 2022, the team assessed the first 2 prototypes at the Valcartier, Quebec, military base.

"Our results clearly disrupted the status quo, which depended on image banks to identify drones," says Guillaume Gagné, Defence Scientist at the DRDC Valcartier Research Centre. "The test showed that this lightweight technology can be housed in a small box, consume very little power and—most importantly—guarantee excellent accuracy in a visually congested environment."

While the technology is not yet scalable for commercial use, the multi-talented research team is working on modifications that will take it to the next level.
"We're developing the next generation of the prototype that has recently been tested in collaboration with the DRDC and the Canadian Armed Forces," says Marc-Antoine Drouin, Senior Research Officer with the NRC's Computer Vision and Graphics team. "We expect to double the detection range, add a radio link to communicate detection and connect the system to command and control software—allowing full integration into the drone-detection ecosystem." The tests will pinpoint missing features or limitations that need to be addressed before it can be market-ready.

This success aligns with the DRDC's mandate to develop technology in support of the CAF's operational needs. "We are crafting a road map for a multi-year project with the NRC's Digital Technologies Research Centre to continue advancing the current prototype using emerging approaches and technologies," adds Guillaume.

He also points out that the 3-year project builds on the collaboration between these two government entities that goes back more than 75 years. This history creates the ideal partnership to support national defence and align military and civil security. Their continuing inventiveness will contribute to Canada's national security well into the future.

For a more in-depth look at the research behind the story, read the team's recent article, A Virtual Fence for Drones: Efficiently Detecting Propeller Blades with a DVXplorer Event Camera.

Contact us
Media Relations, National Research Council of Canada
1-855-282-1637 (toll-free in Canada only)
1-613-991-1431 (elsewhere in North America)
001-613-991-1431 (international)
media@nrc-cnrc.gc.ca
Follow us on Twitter: @NRC_CNRC

———————————————————————————-
The above link is the article’s abstract and in turn links to the PDF that I referred to in my second paragraph:

https://dl.acm.org/doi/pdf/10.1145/3546790.3546800

I very much like this! Good sleuthing. Happy weekend all!!

 

[Damo4]

 

[Xray1]

Akida can be used wherever sensor output classification is required.

The principles underlying Akida can be used all the way up to the cloud.

BRN chose the edge because that was where there was the least competition and the largest potential market. The closer you get to the cloud, the smaller the potential market in terms of the number of chips.

So, yes, Dell edge servers are a candidate.

Late edition: In fact, with the problem of climate change, there should be a legislative requirement for Akida all the way through to the cloud.

Like what you posted. Especially the part about:

"Late edition: In fact, with the problem of climate change, there should be a legislative requirement for Akida all the way through to the cloud."

I personally think that BRN should also seriously consider engaging the services of a Co that can provide advise & deal in "Carbon Credits" as potentially another source of revenue for the Co, given the overall low power requirements of our technology.

 

[Sirod69]

also on LinkedIn

BrainChipBrainChip


BrainChip will participate in the Global Leadership Summit. This invitation-only exclusive event brings together the leading global C-level and senior executives from the semiconductor and high technology industry. https://lnkd.in/gZmDrf2u

 

[Deleted member 118]

There must be more videos from the event as I can’t find any apart from this

 

[Xray1]

also on LinkedIn

BrainChipBrainChip

BrainChip will participate in the Global Leadership Summit. This invitation-only exclusive event brings together the leading global C-level and senior executives from the semiconductor and high technology industry. https://lnkd.in/gZmDrf2u

IMO ........... Great to see some other Big named Co's attending and should also be a good opportunity for a bit of Co networking with the other attendee's.

 

[Tothemoon24]

LANL Researchers Design New Artificial Synapses for Neuromorphic Computing​

June 2, 2023

June 2, 2023 — The human brain has been called the most complicated object in the universe. Trying to replicate that still-unmatched capability for computing, scientists at Los Alamos National Laboratory have made a new interface-type memristive device, which their results suggest can be used to build artificial synapses for next-generation neuromorphic computing. Memristive devices, or memristors, represent long-sought circuit technology that, unlike current resistor technology, has both programming and memory capabilities — memristors could remember which electrical state they were in when powered off, a human brain-like ability that opens up new possibilities for computing and devices.
Tested against a dataset of handwritten images from the Modified National Standards and Technology database, the interface-type memristors realized a high image recognition accuracy of 94.72%. Image: LANL.
“Data processing is an essential part of today’s science, with machine learning, artificial intelligence and artificial neural networks used to address pressing questions in everything from climate science to national security applications,” said Aiping Chen, Laboratory scientist with the Center for Integrated Nanotechnologies. “But conventional computing architecture demands a great deal of energy and is increasingly less able to scale up to meet bigger and bigger data challenges. Neuromorphic computing, which mimics the unmatched data storage and processing architecture and capabilities of the human brain, offers a path to continue to extend computing performance.”
Conventional computing is constrained by the so-called von Neumann bottleneck, in which computing and memory are separate. Processing advanced tasks like machine learning and image recognition on digital computers consume a significant amount of energy and time due to transferring the data back and forth between a central processing unit and memory. Data center energy consumption has increased rapidly in the past few years, with projections that approximately 8% of the world’s electricity will be used by data centers by 2030.
Additionally, in conventional computer architecture, billions of transistors on silicon-based microchips serve as switches for a computer’s binary code. Physical limits to the miniaturization of those transistors have helped spell the end of Moore’s Law, a maxim that foretold the doubling of processing power roughly every two years.
In-Memory Computing: Just Like a Brain
Co-locating information storage and processing at synapses, which connect the 100 billion neurons sending and receiving chemical information, the human brain’s “in-memory processing” saves time and energy. Neuromorphic computing relies on emergent devices such as memristors, switches between two terminals that control and remember the charge flowing through, to replicate the structure and function of synapses.
In the fast-evolving field of neuromorphic computing, memristor designs have included filament systems, in which a charge is delivered through the devices. But, prone to overheating, filament systems lack stability and reliability.
Chen and his colleagues are working on a different approach called an interface-type memristor, and have produced a reliable, high-performing device with a simple structure based on an Au/Nb-doped SrTiO3 interface — essentially gold and other semiconducting materials. The interface-type memristors can, in principle, be scaled down to nanometer size that even filament-based memristor technology cannot achieve. (By contrast, a human hair is approximately 100,000 nanometers thick.) And especially in contrast to transistor-based neuromorphic chips, the interface-type memristive device needs significantly less power to fuel its processing.
“Different from digital computing with a von Neumann architecture, neuromorphic computing, inspired by biological systems, works just like a brain,” said Chen. “The advantages of that structure include low-energy consumption, high parallelism and excellent error tolerance. The human brain runs at only 20 watts, after all, but learns extremely effectively. These advantages make it very good for advanced computing tasks like learning, recognition and decision-making.”
Excelling at Advanced Computing Tasks
The team used artificial neural-network simulation to study the computing performance of the interface-type memristor, testing it against a dataset of handwritten images from the Modified National Standards and Technology database maintained by the National Institute of Standards and Technology. Demonstrating excellent uniformity, programmability and reliability, the device realized a recognition accuracy of 94.72%.
That performance makes the team believe these new interface-type memristive devices can be a fundamental hardware piece for next-generation neuromorphic computing.
“The capabilities we’re seeing suggest that neuromorphic chips, like human brains, will be good at advanced tasks that include learning and real-time decision-making,” said Chen. “We could see neuromorphic computing enable a lot of applications that require intelligence, from self-driving cars, to drones, to security cameras. Basically, many things that people are capable of doing, these types of devices will be able to do.”
The team plans to continue to develop the technology with an emphasis on the need for co-design — hardware design informed by algorithmic approaches offered by computer scientists.
Paper: “An Interface-Type Memristive Device for Artificial Synapse and Neuromorphic Computing,” Advanced Intelligent Systems. DOI: 10.1002/aisy.202300035

 

[IloveLamp]

Watch the video..........

Automotive News Video: Magna’s in-mirror monitoring system Video from Automotive News

Magna’s in-mirror monitoring system tracks the driver’s head, eye and body movement to help detect signs of distracted behavior, drowsiness and fatigue.

hubs.ly

 

[Deleted member 118]

 

[Deleted member 118]

 

[Tuliptrader]

Would have required a ASX notification for the remainder of the Capitol Call, ............................ i would have thought.

also on LinkedIn

BrainChipBrainChip

BrainChip will participate in the Global Leadership Summit. This invitation-only exclusive event brings together the leading global C-level and senior executives from the semiconductor and high technology industry. https://lnkd.in/gZmDrf2u

Have a look a the absolute calibre of the companies at the bottom of the page. And there we are smack bang in the middle amongst these titans. How did we get an invite to this? Not bad for meme stock,eh.

GSA Lightning Community

TT

 

[Deleted member 118]

 

[Deleted member 118]

 

[Deleted member 118]

 

[Deleted member 118]

 

[Deleted member 118]

 

[TopCat]

Renesas Completes Acquisition of Panthronics2. June 2023​

 

[Deleted member 118]

I tried to pick out a few videos this morning that might be off interest, but not watch any so apologie if irrelevant.

 

[supersonic001]

BrainChip on LinkedIn: LT Event

BrainChip will participate in the Global Leadership Summit. This invitation-only exclusive event brings together the leading global C-level and senior…

www.linkedin.com