Self-Healing Visual Recovery for Autonomous Ground Vehicles Using Camera-Only Visual Odometry

2026-07-13

AI Summary

This study addresses the challenge of sustained lane-following for low-cost monocular-camera-equipped unmanned ground vehicles when the guidance line is occluded or lost. The authors propose a lightweight, two-stage self-recovery method: first attempting to reacquire the line through in-place rotation and dynamically relaxed HSV color thresholds; if unsuccessful, the system leverages monocular visual odometry to backtrack to a previously saved “breadcrumb” waypoint and restart tracking. Notably, this approach achieves closed-loop, self-healing visual recovery without LiDAR, GPS, or GPU support, embedding a complete MAPE-K control loop within a 50 ms cycle and operating entirely on CPU at 20 Hz. Experimental results in Webots across three tracks with 119 injected failures demonstrate an 86.6% successful recovery rate and a median recovery time of 3.26 seconds.

Abstract

Low-cost unmanned ground vehicles are often used in indoor places like warehouses, inspection corridors, and farm rows, where painted floor lines guide the robot. Line following is useful because it only needs one camera and little computing power, but it can fail when the line is blocked or turns sharply and goes out of view. Sensor-rich platforms tolerate this through hardware redundancy (LiDAR, GPS, multiple cameras), but camera-only systems must recover at runtime with no additional infrastructure. This paper presents a lightweight, two-stage recovery approach that restores guideline tracking without LiDAR, GPS, or a GPU. When the line is lost, the robot first turns in place while slowly relaxing its color checks and waiting for confirmation across multiple frames (Stage 1). If the line is still not found, monocular visual odometry moves the robot back to saved breadcrumb positions before it tries again (Stage 2). The system uses a depth-gated HSV line tracker, a YOLOv8n obstacle detector, and a visual odometry breadcrumb mapper, and it runs at 20 Hz on CPU-only hardware. The controller embeds a complete MAPE-K loop within a single 50 ms control tick, with no external adaptation manager required. The approach is evaluated across 119 fault-injected episodes on three Webots simulation courses. The method was successful in 86.6% of cases, with a median recovery time of 3.26 seconds. These results demonstrate that reliable visual recovery is feasible on camera-only UGVs within practical cost and computational limits.

Problem

• visual recovery
• camera-only navigation
• line following failure
• autonomous ground vehicles
• visual odometry

Innovation

• visual odometry
• self-healing navigation
• camera-only UGV
• MAPE-K loop
• breadcrumb mapping