The global COVID-19 pandemic has inevitably made disinfection a daily routine to ensure the safety of public and private spaces. However, the existing disinfection procedures are time-consuming and require intensive human labor to apply chemical-based disinfectant onto contaminated surfaces. In this paper, a robot disinfection system is presented to increase the automation of the disinfection task to assist humans in performing routine disinfection safely and efficiently. This paper presents a semi-autonomous quadruped robot called LASER-D for performing disinfection in cluttered environments. The robot is equipped with a spray-based disinfection system and leverages the body motion to control the spray action without an extra stabilization mechanism. The spraying unit is mounted on the robot’s back and controlled by the robot computer. The control architecture is designed based on force control, resulting in navigating rough terrains and the flexibility in controlling the body motion during standing and walking for the disinfection task. The robot also uses the vision system to improve localization and maintain desired distance to the disinfection surface. The system incorporates image processing capability to evaluate disinfected regions with high accuracy. This feedback is then used to adjust the disinfection plan to guarantee that all assigned areas are disinfected properly. The system is also equipped with highly integrated simulation software to design, simulate and evaluate disinfection plans effectively. This work has allowed the robot to successfully carry out effective disinfection experiments while safely traversing through cluttered environments, climb stairs/slopes, and navigate on slippery surfaces.