Indoor navigation within commercial animal farming houses is crucial for the increasing operations carried out by mobile robots. However, existing methods often suffer from issues such as sensitivity to ambient light, insufficient accuracy for precise animal operations, or complexity in implementation. In this study, we proposed a novel navigation method for mobile chassis in a commercial stacked cage farming house, utilizing an active light source, self-designed fiducial marker, and machine vision. Our navigation system incorporated imaging modules equipped with 850 nm infrared light to minimize animal stress and mitigate the effects of low illumination. The self-designed fiducial marker, DRITag, enabled precise maker identification with a recognition rate of 100 % and achieved high positioning accuracy, with an absolute mean error of 0.34 cm within a range of 1.2 m. To determine the chassis’ yaw angle, we employed the U-net deep learning network, which achieved satisfactory root mean square errors of 0.22° and 0.10° on the ground and upper floors, respectively. Global localization of the chassis across different farm areas was achieved through a combination of DRITag and odometry information. Our proposed method provided a flexible and user-friendly approach to navigation path planning, employing a waypoint task text file that enabled easy modification of the path through simple text file editing. Field tests conducted in a two-story commercial stacked cage chicken farming house demonstrated the high accuracy of our navigation method, with standard deviations of longitudinal deviations below 1.85 cm and 1.82 cm, and standard deviations of lateral deviations below 1.50 cm and 1.12 cm at a speed of below 0.3 m/s on the ground floor and at a speed of 0.2 m/s on the upper floor, respectively. Overall, our study presented an independent-of-ambient-light, highly accurate, and easily implementable navigation method for mobile chassis in commercial chicken farming houses, with significant potential for future research in mobile robotics within animal farming environments.