What Matters in Constructing a Visual Servoing Scheme: A Review of Key Issues and Solutions

Visual servoing (VS) leverages vision as a perception tool to compute robot control inputs based on vision feedback, enabling real-time sensorimotor control. Its key advantages include high control accuracy with simple camera systems, cost-effectiveness, and scalability. As an enabling technology, VS plays a critical role in advancing robot applications. While existing reviews explore VS’s current state and future prospects in medical, agricultural, and industrial sectors, challenges remain. These include ensuring stability and convergence in diverse dynamic systems. This paper concentrates on these general and specific technical challenges, providing a problem-oriented overview to guide researchers in applying VS effectively.