TinyML & Arduino in AI & Artificial Intelligence

Deploying AI to a Microcontroller (MCU) is a battle against physics. Unlike a server with gigabytes of RAM, an MCU might only have 256KB of SRAM and 1MB of Flash. We cannot use standard TensorFlow or even standard TFLite. Instead, we use TFLite Micro (TFLM), a specialized runtime that uses Zero Dynamic Allocation. This means it never calls malloc() or new, preventing unpredictable memory crashes in critical embedded systems.

Because TFLite Micro doesn't use dynamic memory, the developer must provide a Tensor Arena— a static byte array in SRAM. During the AllocateTensors() phase, the interpreter maps every tensor in the model's graph into this buffer. If the arena is too small, the code will fail to boot. If it's too large, you won't have room for your sensor data or Wi-Fi stack. Balancing the arena size is the core challenge of TinyML engineering.

To save precious RAM, we store the model's weights in Flash Memory (Read-Only) using the PROGMEM attribute. The weights are accessed directly from Flash during inference. The SRAM (RAM) is reserved only for the intermediate 'Activations'—the temporary mathematical results that change as data flows through the layers. This split architecture allows us to run 1MB models on devices with only 100KB of working memory.

Welcome to TinyML! We are bringing machine learning to tiny, low-power microcontrollers like the Arduino Nano 33 BLE Sense.

These devices have constraints. An Arduino might only have 256KB of RAM. Standard TensorFlow models are megabytes or gigabytes in size. We must adapt.

The solution? TensorFlow Lite for Microcontrollers (TFLM). It runs a minimal interpreter written in C++11, allocating zero dynamic memory.

Checkpoint: Why can't we run standard TensorFlow models directly on an Arduino?

• →Insufficient RAM/Flash memory
• →Arduino doesn't support C++

To run a model in C++, we convert the .tflite file into a C byte array. This array is stored in Flash memory using the PROGMEM keyword.

In the setup() function, we instantiate the micro interpreter. We pass it the model, operations resolver, and the memory arena we carved out.

Checkpoint: Which function maps the active tensors into the pre-defined tensor arena?

• →AllocateTensors()
• →Invoke()

TinyML workflow mastered! You've learned to bridge AI and embedded C++. Ready to explore microcontroller deployment?

Check a Real Microcontroller Memory Budget. Finish checking whether a model fits inside a microcontroller's tiny memory budget.