RobjHunt
Regular
I like this one better.
That was absolutely gold!!
I like this one better.
Back in the mid-90s, Australia was a leader in solar research. I went to a seminar with Howard's chief scientist and, instead of advocating for the support of such industries, he parroted the economic rationalist kant that government should not be in the business of picking winners. Shortly thereafter, the Australian technology was in production in China under the control of the then richest man in the world.I heard a while back that Germany has the highest use of solar power per capita than any other country - despite their climate. Not sure if that's still the case but the fact that it isn't Australia, by a long way, is just shameful I reckon. So far we suck at being green......
![]()
Infineon's Smart Alarm System Puts Machine Learning to Bear on Breaking Glass and Other Triggers
Offering a claimed five year runtime on battery alone, this edge AI pops a clever sensor fusion algorithm on a low-power SoC.www.hackster.io
"Infineon Technologies has launched what it claims to be the "industry's first" artificially intelligent acoustic event and sensor fusion alarm system to be driven wholly by battery power: the Smart Alarm System, or SAS.
"We are excited to enable a unique and differentiated approach to bringing AI/ML [Artificial Intelligence/Machine Learning] capabilities to cost-sensitive, battery-powered home security sensor systems, without sacrificing battery life," says Infineon's Laurent Remont of the launch.
"Current home security solutions are unreliable for detecting events such as glass break[ing]," Remont continues. "Our new solution combines a number of best-in-class technologies to create an alarm system that is smart, reliable and power efficient. We look forward to bringing more innovative solutions into the home security market."
The alarm system is built around Infineon's PSoC 62 "Performance Line" system-on-chip, which offers two Arm Cortex-M4 cores and an Arm Cortex-M0+ core — enough power to run an on-device machine learning algorithm that can differentiate between alarm-triggering events and the normal sounds of a home or office, yet to do so efficiently enough to allow the system to run entirely on battery for up to five years.
Elsewhere on the board is an XENSIV MEMS microphone and a digital pressure sensor, with data from both being fed into the on-board ML algorithm to avoid false positives triggering the alarm unnecessarily. In addition to triggering on the sound of breaking glass, indicating an intruder has broken a window to gain entry, the alarm system can also send out alerts when less-smart audible devices like smoke or carbon monoxide alarms have triggered.
A reference design incorporating all these features is available to Infineon customers now, with a physical board shipping in September this year. The company has also partnered with Flex to offer the same capabilities in a system-in-package (SIP) format. Additional information is available on the official product page."
Interesting...
Sounds like a great trip.So much to catch up on, ( maybe 40-50 pages) after only now reaching almost the half way point of my holidays. Currently in Zanzibar and heading to Kenya and Tanzania next to go to the Serengeti. Definitely better than being back in Melbourne View attachment 10507
I saw that too. I became aware because they also use the Arm Cortex-M4 like we do for our PCI board. Then the statement that the battery lasts for years and "machine learning Tiny AI algorithms".![]()
Infineon's Smart Alarm System Puts Machine Learning to Bear on Breaking Glass and Other Triggers
Offering a claimed five year runtime on battery alone, this edge AI pops a clever sensor fusion algorithm on a low-power SoC.www.hackster.io
"Infineon Technologies has launched what it claims to be the "industry's first" artificially intelligent acoustic event and sensor fusion alarm system to be driven wholly by battery power: the Smart Alarm System, or SAS.
"We are excited to enable a unique and differentiated approach to bringing AI/ML [Artificial Intelligence/Machine Learning] capabilities to cost-sensitive, battery-powered home security sensor systems, without sacrificing battery life," says Infineon's Laurent Remont of the launch.
"Current home security solutions are unreliable for detecting events such as glass break[ing]," Remont continues. "Our new solution combines a number of best-in-class technologies to create an alarm system that is smart, reliable and power efficient. We look forward to bringing more innovative solutions into the home security market."
The alarm system is built around Infineon's PSoC 62 "Performance Line" system-on-chip, which offers two Arm Cortex-M4 cores and an Arm Cortex-M0+ core — enough power to run an on-device machine learning algorithm that can differentiate between alarm-triggering events and the normal sounds of a home or office, yet to do so efficiently enough to allow the system to run entirely on battery for up to five years.
Elsewhere on the board is an XENSIV MEMS microphone and a digital pressure sensor, with data from both being fed into the on-board ML algorithm to avoid false positives triggering the alarm unnecessarily. In addition to triggering on the sound of breaking glass, indicating an intruder has broken a window to gain entry, the alarm system can also send out alerts when less-smart audible devices like smoke or carbon monoxide alarms have triggered.
A reference design incorporating all these features is available to Infineon customers now, with a physical board shipping in September this year. The company has also partnered with Flex to offer the same capabilities in a system-in-package (SIP) format. Additional information is available on the official product page."
Interesting...
I saw that too. I became aware because they also use the Arm Cortex-M4 like we do for our PCI board. Then the statement that the battery lasts for years and "machine learning Tiny AI algorithms".
Is a board only with Arm Cortex-M4s capable of machine learning? Is there a way to see if our IP is in there or not? It should be available in Q3.
"The AI/ ML algorithm is capable of filtering out background noises, such as, baby crying or dog barking and other events that could generate a false positive." This could perhaps take place in their Acoustic Event Detector AED?
Maybe the 10000 eyes can look at this more closely.
https://www.infineon.com/cms/en/app...olutions/alarm-system/?redirId=197381#!boards
Quite possibly not fake. The lady obviously has no clue but the owner of the new Tesla looks to be driving on a fairly remote part of the country?
No electric charge stations anywhere. So, its makes perfect sense (to him)to carry a generator in case you run out. Only problem was he was too dumb to have it filled up.![]()
I think we can thank the last government slowMo and Co for that…I heard a while back that Germany has the highest use of solar power per capita than any other country - despite their climate. Not sure if that's still the case but the fact that it isn't Australia, by a long way, is just shameful I reckon. So far we suck at being green......
So does that mean the ARM chips will potentially come with Akida Integrated?
I think Aust has more roof solar per capita but doesn't have the massive solar farms. Why I always thought Aust was the perfect place for EV's but in the western world we're the worst in ownership.Back in the mid-90s, Australia was a leader in solar research. I went to a seminar with Howard's chief scientist and, instead of advocating for the support of such industries, he parroted the economic rationalist kant that government should not be in the business of picking winners. Shortly thereafter, the Australian technology was in production in China under the control of the then richest man in the world.
That tablecloth is spectacular!Btw who the 2 BrN staff in the photo. This photo was recently taken at the Autosens event in Detroit
View attachment 10464
This is a great video.
“standardised chip that will do all by 2024”