SoC development kits boost endpoint AI and Machine Learning efforts

QuickLogic Corporation, a developer of ultra-low power multi-core voice-enabled SoCs, embedded FPGA IP, and Endpoint AI solutions, announced the launch of its Merced EOS S3AI Hardware Development Kit (HDK) to enable quick development of applications for endpoint Artificial Intelligence (AI) and Machine Learning (ML).

The Merced HDK is an evaluation platform for QuickLogic’s EOS S3AI SoC and includes continuous (time-series) sensors accelerometer, gyroscope, magnetometer, and two digital microphones (PDM). It’s an Arduino form-factor compatible development kit which makes it easy for developers to ramp up an AI design by easily adding different sensors or 3rdparty modules. The HDK links directly with the SensiML Data Analytics Toolkit (free 90-day evaluation period included) which enables a seamless flow from sensor data collection all the way through automatic feature extraction and classifier development using integrated machine learning technology. The AI model binary file is directly programmed by the SensiML Toolkit into the EOS S3AI device on the HDK. The board can easily be updated via an on-board USB-2 UART connection to create demos and prototypes without additional time or resources. Additional sensors can be added to the HDK by way of the available connectors and accompanying SDK.

The Merced EOS S3AI HDK (QAI-EVALKIT-AA-1.0), EOS S3AI SoC (EOSAI-S-PDN64 or EOSAI-S-WRN42) and the SensiML Analytics Toolkit are all available now.