Reliable weld seam perception remains challenging in industrial environments, where arc light, spatter, smoke, and varying seam geometries can seriously degrade visual sensing. These disturbances make it difficult to achieve a unified representation, accurate localization, and real-time inference at the same time. To address this problem, this paper presents an end-to-end lightweight framework for weld seam keypoint detection and tracking based on an improved SimCC. A unified three-keypoint formulation is introduced to represent different weld geometries by using one seam center point and two orientation reference points, thereby supporting a perception-to-control mapping in which position control and orientation control are decoupled. In addition, a lightweight C3k2-based backbone is designed, and a non-parametric log-domain quadratic peak-refinement decoder is proposed to alleviate the discretization-induced quantization error of SimCC classification distributions without adding model parameters. Experiments show that the proposed model contains only 1.4 M parameters, achieves 17.01 ms CPU inference latency, and obtains a detection accuracy of 1.89 px MAE. In curved weld seam tracking experiments with the integrated robotic system, it further achieves an average trajectory tracking error as low as 0.159 mm and an average orientation error of 3.738°, demonstrating its real-time accuracy and robustness for industrial welding applications.

This study proposes a hierarchical drive control system to ensure speed stability in dual-motor tracked vehicles operating under complex terrain and heavy-load conditions. The system adopts a two-layer structure. At the upper level, the sliding mode controller is designed for both longitudinal speed regulation and yaw rate control, thereby stabilizing straight line motion and the steering maneuvers. At the lower level, a synchronization mechanism aligns the velocities of the two motors, enhancing the vehicle’s robustness against speed fluctuations. Simulation results demonstrate that, across both heavy load and light load bulldozing scenarios, the deviation between the controller output and the reference command remains within 5$\%$. These findings confirm the accuracy of the control implementation and validate the effectiveness of the proposed framework.