Robust Visual Servoing for Mobile Robots Under Illumination Variance: A Domain Adaptation Approach

Visual servoing offers a powerful approach for controlling robot motion using visual feedback. However, performance degrades significantly under varying illumination conditions common in real-world deployments. This paper investigates the application of domain adaptation techniques to improve the robustness of visual servoing systems to illumination variance. We propose a novel approach that leverages synthetic data and unsupervised learning to train a robust visual feature extractor, enabling accurate and reliable visual servoing in challenging lighting environments. Experimental results demonstrate the effectiveness of our proposed method in achieving superior tracking accuracy and convergence compared to traditional visual servoing approaches under significant illumination changes.