Integrated YOLOP Perception and Lyapunov-based Control for Autonomous Mobile Robot Navigation on Track

TL;DR

This paper introduces a real-time autonomous navigation system for mobile robots that combines advanced perception techniques with a Lyapunov-based control method, ensuring stable and accurate lane following without reliance on prior maps.

Contribution

The work integrates a multi-task perception pipeline with a Lyapunov-based controller for autonomous navigation, providing stability guarantees and real-time performance in dynamic environments.

Findings

Successful real-world implementation on embedded platforms

Stable and smooth trajectory tracking demonstrated

No dependence on HD maps or global localization

Abstract

This work presents a real-time autonomous track navigation framework for nonholonomic differential-drive mobile robots by jointly integrating multi-task visual perception and a provably stable tracking controller. The perception pipeline reconstructs lane centerlines using 2D-to-3D camera projection, arc-length based uniform point resampling, and cubic polynomial fitting solved via robust QR least-squares optimization. The controller regulates robot linear and angular velocities through a Lyapunov-stability grounded design, ensuring bounded error dynamics and asymptotic convergence of position and heading deviations even in dynamic and partially perceived lane scenarios, without relying on HD prior maps or global satellite localization. Real-world experiments on embedded platforms verify system fidelity, real-time execution, trajectory smoothness, and closed-loop stability for reliable autonomous navigation.