<p>Inspection and maintenance of large structures, such as ship hulls and oil tanks, are essential for ensuring safety and operational efficiency. Traditional manual inspection methods are often labor-intensive, time-consuming, and potentially hazardous. This study proposes a novel solution: a lizard-inspired quadruped wall-climbing robot (LQWCR). Inspired by the morphology and locomotion of lizards, the robot’s leg design, kinematics, and dynamics are systematically analyzed. To facilitate control, the system dynamics are linearized using a Back Propagation (BP) neural network model. Based on this model, innovative motion control strategies are developed to achieve precise trajectory tracking. Additionally, compliance control strategies are introduced to mitigate impact forces when the robot’s foot interacts with welds, improving its adaptability to unstructured environments. Simulation results demonstrate that these strategies effectively enable the robot to negotiate weld seams and reduce the risk of falling. This research lays a theoretical foundation and provides technical support for applying wall-climbing robots to inspection, maintenance, and surveillance tasks in real-world scenarios.</p>