Quickso AutoML, the 'TinyML' model acceleration platform
ST MLC Sensors Now Available in Quickso AutoML


Quixo, the developer of the Quixo AutoML platform that accelerates the development of ultra-small machine learning (TinyML) models for edge computing, and STMicroelectronics announced on the 15th that they will support the use of ST Machine Learning Core (MLC) sensors in Quixo AutoML.

ST MLC sensors run sensing algorithms implemented on the sensed data set locally instead of on the host processor, reducing overall system power consumption. Quickso AutoML automatically generates machine learning solutions optimized for edge devices with ultra-low latency, ultra-low power requirements, and a very small memory footprint using this sensor data.

These algorithmic solutions overcome die size limitations due to computing power and memory size, and extend system battery life by tailoring machine learning models to the sensors.

“Application developers can now quickly implement and deploy machine learning algorithms on ST MLC sensors without consuming MCU cycles or system resources across a wide range of applications, including industrial and IoT use cases,” said Lee Sang-won, CEO of Quickso.

“Placing MLC in a sensor, such as the ST LSM6DSOX or ISM330DHCX, can reduce system data transmission and cut system power consumption by a factor of 10,” said Simone Ferri, ST’s Business Director of MEMS Sensors. “It also provides enhanced event detection performance, wake-up logic, and real-time edge computing capabilities.”