Supported Platforms

Broad IoT Edge Platform Support

As an AutoML toolkit for IoT edge devices, SensiML covers a broad array of embedded endpoint system architectures enabling developers to target the right system for their application needs. From small ultra-low power MCUs possessing tens of kB memory to multi-core x86 client nodes with GBs of SRAM, SensiML provides ML code output to suit the hardware capabilities of your chosen platform.

Architecture supported include:

SensiML supports multiple levels of platform integration. The maximum flexibility is afforded by Standard and Enterprise versions of SensiML that offer the option of generating full C source code output. This gives the developer full freedom to compile the Knowledge Pack in their own development environment for their custom target device using training data gathered by user-means and imported as a dataset into SensiML.

The next form of integration is awareness and support of unique hardware capabilities allowing for utilization of specific SoC accelerators or compute cores for those devices supporting this capability. Examples of this level of hardware support include DSP and FPGA optimizations for Quicklogic EOS S3 and CMSIS DSP optimization for Arm Cortex processor based SOCs.

The next form of platform support includes specific evaluation and prototyping boards that are recognized directly within the SensiML Data Capture Lab (DCL) for interactive data acquisition of raw sensor data into the toolkit without need to utilize custom scripts, 3rd party data acquisition software, or custom coding. Boards supported directly in DCL include STMicro STM32 (ST SensorTile), Nordic nRF52 (Nordic Thingy52), and QuickLogic EOS S3 (QuickAI), with more soon to be announced.

Finally, for user-designed boards and those having unique sensors, we provide an integration guide illustrating the steps to add full integration to SensiML Toolkit for custom hardware of your own.

QuickLogic – Quick AI Accelerated AI Platforms

Unique ultra-low power AI platforms with heterogeneous processing cores, the QuickLogic family of endpoint AI platforms based on the EOS S3 SoC, allow SensiML algorithms to be implemented with hardware optimization for its included Arm Cortex M4F, FPGA fabric, and DSP compute cores.

A Range of Hardware Architectures

For the target hardware SensiML generates optimized code which takes advantage of DSP, FGPA and other device hardware for accelerating the data processing and classification pipeline.

Download

ST – STM32 & SensorTile Development Kit

The SensorTile is a tiny, square-shaped IoT module that packs powerful processing capabilities leveraging an 80 MHz STM32L476JGY microcontroller and Bluetooth low energy connectivity based on BlueNRG-MS network processor as well as a wide spectrum of motion and environmental MEMS sensors, including a digital microphone.

NXP i.MX RT1050 Crossover MCU and i.MX RT1050 Evaluation Kit

The i.MX RT series of crossover MCUs combine high performance and real time functionality with integration, security and MCU-level usability at an affordable price.

Nordic Semiconductor nRF52 Nordic Thingy IoT Sensor Kit

Nordic Thingy:52® is natively supported within SensiML Analytics Studio

Intel Atom E Processors

Built for embedded applications that power mobile, portable, and small-scale devices of every kind.

ARM Cortex-M Series Processors

ARM Cortex-M3 and higher processors are supported within SensiML Analytics Studio

Raspberry Pi 3/3Bs Supported within SensiML Analytics Studio

The Raspberry Pi 3 Model B+ is the latest product in the Raspberry Pi 3 range, boasting a 64-bit quad core processor running at 1.4GHz, dual-band 2.4GHz and 5GHz wireless LAN, Bluetooth 4.2/BLE, faster Ethernet, and PoE capability via a separate PoE HAT.