Artificial intelligence (AI) requires extreme computational horsepower, but Maxim is cutting the power cord from AI insights. The MAX78002 is a new breed of AI microcontroller built to enable neural networks to execute at ultra-low power and live at the edge of the IoT. This product combines the most energy-efficient AI processing with Maxim's proven ultra-low power microcontrollers. Our hardware-based convolutional neural network (CNN) accelerator enables battery-powered applications to execute AI inferences while spending only millijoules of energy.
The MAX78002 is an advanced system-on-chip featuring an Arm® Cortex®-M4 with FPU CPU for efficient system control with an ultra-low-power deep neural network accelerator. The CNN engine has a weight storage memory of 2MB, and can support 1-, 2-, 4-, and 8-bit weights (supporting networks of up to 16 million 1-bit weights). The CNN weight memory is SRAM-based, so AI network updates can be made on the fly. The CNN engine also has 1.3MB of data memory. The CNN architecture is highly flexible, allowing networks to be trained in conventional toolsets like PyTorch®and TensorFlow®, then converted for execution on the MAX78002 using tools provided by Analog Devices.
In addition to the memory in the CNN engine, the MAX78002 has large on-chip system memory for the microcontroller core, with 2.5MB flash and up to 384KB SRAM. Multiple high-speed and low-power communications interfaces are supported, including I2S, MIPI CSI-2®, parallel camera interface (PCIF) and SD3.0/SDIO3.0/eMMC4.51 secure digital..
The device is available in a 144-pin CSBGA (12mm x 12mm, 0.8mm pitch) package.
The MAX78002 evaluation kit (EV kit) provides a platform and tools for leveraging device capabilities to build new generations of artificial intelligence (AI) products.
The kit provides optimal versatility with a modular peripheral architecture, allowing a variety of input and output devices to be remotely located. DVP and CSI cameras, I2S audio peripherals, digital microphones, and analog sensors are supported, while a pair of industry-standard QWIIC connectors supports a large and growing array of aftermarket development boards. An onboard stereo audio codec offers line-level audio input and output, and tactile input is provided by a touch-enabled 2.4in TFT display. The MAX78002 energy consumption is tracked by a power accumulator, with four channels of formatted results presented on a secondary TFT display. All device GPIOs are accessible on 0.1in pin headers. A standard coaxial power jack serves as power input, using the included 5V, 3A wall-mount adapter. Two USB connectors provide serial access to the MAX78002, one directly and the other through a USB to UART bridge. A third USB connector allows access to the MAX78002 energy consumption data. Rounding out the features, a microSD connector provides the capability for inexpensive highdensity portable data storage.
Applications
ANStek|Analog Devices Microsite:https://macnica-adi.com
Line@:https://lin.ee/5gcKNi7
Facebook:https://www.facebook.com/ANStek3528
Sample apply & Technical Support:https://www.surveycake.com/s/dQ3Y2
Mail:marketing@anstek.com.tw