A Vision-Based Four-Cable-Driven Positioning System with Cascade Control: Experimental Evaluation

Effective water quality monitoring in large-scale aquaculture remains a challenge due to the limited spatial coverage of fixed sensors and the operational constraints of Unmanned Aerial Vehicles (UAVs). To address this, this paper presents the design and validation of a scalable four-cable-driven parallel robot (CDPR) with a four-corner architecture tailored for autonomous sensor deployment. The system implements a cascade control strategy, integrating a low-level motor current loop with a high-level vision-based kinematic loop to ensure precise end-effector positioning under dynamic uncertainties. Although the mechanical design supports five degrees of freedom (5-DoF), this study focuses on validating the translational performance required for environmental sensing. Experimental validation was conducted using square and circular trajectories under varying payloads. Results demonstrate that the system maintains a consistently low normalized tracking error even under maximum load conditions, confirming the robustness and feasibility of the proposed architecture as a cost-effective solution for smart aquaculture monitoring.