Currently, Lattice Semiconductor has enhanced its Lattice sensAI solution stack to accelerate AI/ML application development on low-power Lattice FPGas. The new release includes the Lattice sensAI Studio design environment for end-to-end ML model training, validation, and compilation. These improvements include support for the Lattice Propel design environment based on embedded processor development and support for the TensorFlow Lite deep learning framework for reasoning on devices. In sensAI 4.0, developers can use a simple drag-and-drop interface to produce FPGA designs, using risC-V processors and CNN acceleration engines to facilitate quick and easy implementation of ML applications on power-constrained Edge devices.
Ai /ML models can be trained to support a variety of device applications that require low-power operation at the edge, including security and surveillance cameras, industrial robots, consumer robots and toys. The solution stack enables developers to quickly create AI/ML applications running on flexible, low-power dot-array FPgas.
Hideto Kotani, division chief at Canon Inc Said："Lattice's embedded vision low-power FPGA and sensAI solution stack with Edge AI/ML applications play a vital role in helping us bring cutting-edge smart iot devices to market quickly and efficiently."
Hussein Osman, director of marketing for Lattice, said:"With the support of TensorFlow Lite and the new Lattice sensAI Studio, it's now easier than ever for developers to leverage our sensAI stack to create AI/ML applications that can run on battery-powered Edge devices."
