Improve Accuracy and Expand Battery Life for AI powered Cameras

Improve Accuracy and Expand Battery Life for AI powered Cameras Grovety selected Alif Ensemble due to the multi-core architecture with Ethos-U55 and Alif’s aiPM technology™, which allows finite control over power domains.https://www.youtube.com/watch?si=VLOXSndbf3BKtBeQ&v=4saoH3KZGb4&feature=youtu.be Grovety is an embedded AI applications company offering support in firmware, toolchains, and applications. Antony Vasilev, CO-CEO at Grovety, set out on a mission to help the Persian leopards of Romania. After visiting a local national park, Antony discovered that park rangers were struggling to track and protect the endangered Persian leopard population. The park’s wildlife cameras were capturing poor footage and had low accuracy. False positives are not only frustrating, but also drain battery life as the camera is constantly waking up unecessaarily. In this webinar, you will learn how Grovety created and deployed an AI module on Alif Ensemble that improves accuracy, reduces false positives, and expands battery life for wildlife AI cameras. Grovety selected Alif Ensemble due to the multi-core architecture with Ethos-U55 and Alif’s aiPM technology™, which allows finite control over power domains. www.arrow.com

Grovety at tinyML Asia 2023

Grovety at tinyML Asia 2023 November 16, 2023 – At the tinyML Asia 2023 forum in South Korea (Seoul), Anthony Vasilev, Co-CEO of Grovety, a silver sponsor of the forum, showcased new Grovety developments – GRC AI module and GRC Dev Board for review.A self-learning engine doing Anomaly Detection and n-Class Classification in the time series data flow for providing Full turnkey AI functionality for integration in products of various complexity. AI Module is based on the ODL (On-Device Learning) Computing approach for providing and enhancing AI scenarios. (https://github.com/Grovety/grc_sdk/blob/main/docs/GRC_AI-module.md)The GRC Dev Board is a development kit and reference design that simplifies prototyping and testing of advanced industrial AIoT applications such as Anomaly Detection and n-Class Classification for predictive maintenance. (https://github.com/Grovety/grc_devboard/blob/main/docs/ GRC_DevBoard_Description.md)

tinyML Talk

Grovety at tinyML Talk in September 2023 If you missed one of the recent #tinyMLTalks webcast series: Minimizing Resource Usage in Microcontrollers for Cost-Effective Solutions with Ilya Gozman, Senior Fellow, Chief AI Architect at Grovety Inc., hosted by Evgeni Gousev from Qualcomm, we have got you covered! The video is now available on tinyML YouTube channel: https://www.youtube.com/watch?v=rBxpPesd2csWe illustrated how Apache TVM, an open source machine learning compiler, can significantly help reduce device cost and energy consumption. We also discussed our choice of Micro TVM and Arm Ethos-U55 on Alif E5 SoC, detailing their unique advantages and how they align with our goals, and revealing the results of our optimization efforts. Finally, we provided an exciting sneak peek into the future of our TinyML efforts, including upcoming products, our goal to further refine networks while maintaining accuracy, and our innovative strategies for ongoing optimization and development in the TinyML field.

Arm Tech Talk

Arm Tech Talk from Grovety: Saving wildlife by using AI-powered trap camera based on Arm Ethos-U55 You all may now that smart wildlife cameras are used today to preserve natural diversity. Grovety concerned about that and thus gave a talk on how smart wildlife camera traps with Al capabilities powered by the Arm Ethos-U55 are being used in conservation efforts to save the endangered Persian leopards. Here’s the link to the talk: https://www.youtube.com/watch?v=VK_LWwz4mdQ Throughout the presentation, we will cover challenges related to architecture, component selection, suitable neural networks search, and tool choices that were overcame to achieve results under a number of constraints. We look forward to sharing insights and our experience and hearing your thoughts in the comments. Let’s give the Arm ecosystem a head start to implement Al in embedded devices together!

TVM con 2023

Grovety at TVMcon 2023 In March 2023 we gave a short presentation at TVMcon – 2023. – Improvements to Ethos-U55 support in TVM including CI on Alif Semiconductor boards. Our Chief AI Architect gave an overview of the changes we’ve made to the Ethos-U55 integration in TVM, including new operators, new variants of supported operators, and other features. You can see more details at the dedicated TVMcon 2023 link: https://www.youtube.com/watch?v=4hHvXyH-tWQ&list=PL_4zDggB-DBp81G1tAME9r0_P5IY9D700&index=2%20

Generating relocatable code for ARM processors

By upgrading the LLVM compiler, we solved the problem when neither LLVM nor the GCC could create the correct Position Independent Code for Cortex M controllers, with the code running in Flash memory rather than in RAM. Now the binary image of the program can be flashed to an arbitrary address and run from it, without being moved to another place.

How to Inference Your Dragon: Survival Guide for Neurochip Makers

In 2018 we took up our first major contract on developing a software devkit for a neural processing unit. Back in those times, I knew the AI market was rapidly growing and contained hundreds of companies. But I could never imagine that by 2021 another hundred or two chipmakers who developed AI accelerators would emerge, or that we would become Arm AI-partner, or that our projects with neuro-chipmakers would evolve into a separate field. This article is an attempt to formalize the knowledge that we gained from several projects and about a hundred dialogs with chipmakers. At the same time, I would be happy if this article could be helpful to someone.

tinyML Sponsor

Grovety is now a member (Silver Sponsor) of the tinyML Foundation. Founded in 2018 by members from Arm, Google, and Qualcomm, tinyML® Foundation is focused on bringing together embedded hardware, software, and machine learning practitioners to discuss collaborations across these disciplines. Today the tinyML Foundation are over 7000 members in 33 countries and 25 Strategic Partners, leading edge tinyML companies. That makes it one of the fastest-growing ML and AI communities across the globe. tinyML Sponsors List

Tagged tinyML