TinyNav: End-to-End TinyML for Real-Time Autonomous Navigation on Microcontrollers

TL;DR

TinyNav demonstrates that a compact, end-to-end TinyML system can enable real-time autonomous navigation on microcontrollers, making low-cost robotics more accessible by reducing reliance on power-intensive processors.

Contribution

The paper introduces TinyNav, a novel, lightweight neural network architecture optimized for microcontrollers, enabling autonomous navigation with high responsiveness and low latency.

Findings

Achieves 30 ms inference latency on ESP32 microcontroller.

Maintains effective obstacle avoidance with a 23k-parameter model.

Validates spatial awareness through correlation analysis and Grad-CAM.

Abstract

Autonomous navigation typically relies on power-intensive processors, limiting accessibility in low-cost robotics. Although microcontrollers offer a resource-efficient alternative, they impose strict constraints on model complexity. We present TinyNav, an end-to-end TinyML system for real-time autonomous navigation on an ESP32 microcontroller. A custom-trained, quantized 2D convolutional neural network processes a 20-frame sliding window of depth data to predict steering and throttle commands. By avoiding 3D convolutions and recurrent layers, the 23k-parameter model achieves 30 ms inference latency. Correlation analysis and Grad-CAM validation indicate consistent spatial awareness and obstacle avoidance behavior. TinyNav demonstrates that responsive autonomous control can be deployed directly on highly constrained edge devices, reducing reliance on external compute resources.