Research Progress on Autonomous Navigation and Path Planning for Wall-Climbing Robots Inside Large-Scale Pressure Equipment

Internal weld grinding and nondestructive testing of large pressure equipment involve confined spaces, reflective metal surfaces, varying wall curvature, and cable drag, posing challenges to the autonomous navigation and path planning of wall-climbing robots. This paper reviews contact-based, passive-vision, laser-based active-vision, and multisensor-fusion methods for weld-seam tracking. Advances in two-dimensional planar, three-dimensional free-space, curved-surface, and multi-robot cooperative path planning are summarized. Drawing on research and engineering practice involving spherical-tank interiors, the paper discusses line-laser seam tracking, continuous single-robot operation, and partition-based dual-robot coordination for cable-entanglement prevention. Existing systems can perform local automatic seam following and operations within predefined regions, but initial localization, traversal of complex areas, and cable-state assessment still require human intervention. Finally, drawing-driven planning and embodied intelligence are discussed as future development directions.