Edge AI use cases and industry apps (continued)
Welcome again, @HxH to
DevHeads!
Thanks for sharing your interests and you definitely have an outlook of the domains and skills. Sticking to your interest will always help by giving that unfair advantage while you're looking for positions in companies. ML is something that's talked about a lot and coming to the embedded side you may look at places like https://sensiml.com/ and https://www.ti.com/technologies/edge-ai.html for knowing what's hot and where the industry is aligned towards. They have some example projects showcased as well deployed on TI hardware platforms.
Last thing you asked, there certainly are positions in the industry related to these but they as small teams(for example in one of the TI seminars I attended recently, they were showcasing EdgeAI solutions and Mr. Shyam (https://www.linkedin.com/in/shyamjagannathan) was explaining how they're poised to expand with automotive, manufacturing use cases. It's like a high-stake, low-volume skill to architect and build these systems. Several startups are in-line as well.
It's definitely a good to know topic irrespective of what's happening in the industry and working on projects + math related to this will give an edge to one's skillset. You're at the right place - let's keep rolling.
Chris Rogers
SensiML
SensiML -Making Sensor Data Sensible
SensiML is the pioneer of software tools for compact AI processing on IoT endpoints. Build product worthy TinyML code for your IoT sensor app with SensiML.
Advancing intelligence at the edge | TI.com
Scalable and efficient vision processors bring real-time intelligence to smart camera systems
3 Replies
https://www.linkedin.com/posts/narsimh_case-in-point-how-oorja-heat-app-nxp-activity-7147121739179712512-jxY2?utm_source=share&utm_medium=member_desktop
EdgeAI application in automotive for state estimation problem. Having worked on filter based algorithm for battery-pack state estimation, this topic is close to my heart. I had worked on this problem all based on a large pile of collected test data profiles and doing parameter estimation from there.
Curious to see what's happening now with the usage of AI utilities. Seems NXP is supporting this tool on SBC-S32V234 processor hardware platforms(https://www.nxp.com/design/design-center/software/eiq-auto-ml-sw-environment:EIQ-AUTO)
Narsimh Kamath on LinkedIn: Case in Point | How oorja Heat App + NX...
One trend to continue watching out for in 2024 will be AI.
Specifically, edge AI is the deployment of machine learning at the edge.
In the context of…
eIQ® Auto ML Software Development Environment
NXP eIQ Auto is an automotive-grade Deep Learning (DL) toolkit enabling developers to introduce DL algorithms
Did you use a similar neural net approach in your battery pack state estimation work?
No, the one I used was experimental data based and works on mathematical data fitting. It's called layered technique as it depends on test data taken at different SOC values from a cell and then a look-up table is populated for the parameters https://www.mathworks.com/content/dam/mathworks/tag-team/Objects/s/76117-sae-2013-battery-estimation-layered-technique.pdf