Abstract: Unstable depth of the knife penetration is often caused by the uneven surface of the tree bark during natural rubber tapping. In this study, a rubber tapping machine was proposed with fixed profiling depth control using ADAMS dynamic simulation. A profiling depth control was integrated into the moderately irregular bark surfaces. The stable cutting depth was determined to effectively reduce the bark damage. A systematic analysis was made on the anatomical structure of the rubber tree bark. According to the technical requirements of rubber tapping, three key influencing factors on the cutting depth were identified: the curvature radius of the profiling component, the stiffness coefficient of the tension spring, and the stiffness coefficient of the torsion spring. A high-precision three-dimensional model of the rubber tree was constructed using 3D laser scanning and reverse engineering techniques. A dynamic simulation model was developed on the ADAMS platform. A Box-Behnken three-factor, three-level experimental test was carried out with the cutting depth qualification rate as the evaluation index. Response surface optimization was conducted to obtain the combination of the optimal parameters: the profiling component curvature radius of 12.35 mm, tension spring stiffness coefficient of 87.19 N/m, and torsion spring stiffness coefficient of 10.04 N·m/rad, under the qualification rate of 95.33%. Simulation results demonstrated that this configuration significantly enhanced the dynamic response performance of the profiling mechanism. The tapping knife effectively followed the bark surface. The field tests were conducted to verify the accuracy of the simulation. Two major cultivated varieties of the rubber tree were selected under a moderately irregular surface. The field results showed that the average cutting depth qualification rate of the machine equipped with the profiling mechanism was 91.77%, which was improved by 12.51 percentage points. Thereby, both the simulation and the practical reliability were achieved under the optimal parameters. In addition, the better performance was also achieved in the compact structure, low manufacturing cost, and user-friendly operation. Compared with the high-cost self-propelled tapping robots, this machine was more suitable for the large-scale mechanical tapping in small- to medium-sized rubber plantations. Economic analysis demonstrated that an investment payback period of 2–3 years was significantly shorter than that of the 10-year design life, indicating the considerable potential for the widespread application. In summary, the theoretical modeling, simulation optimization, and field validation were combined to systematically solve the problem of the depth control in the fixed-position tapping equipment on irregular surfaces. A practical solution was offered for the profiling depth control and the key equipment in the mechanization of the natural rubber harvesting. Future work can focus on the device’s adaptability to the highly irregular tree morphologies for the corrosion resistance of the key components under high-humidity conditions. The intelligent control and visual recognition can also further improve the environmental adaptability, operational stability, and market competitiveness.