The ability to maintain a stable state in the presence of perturbations, known as robustness, is a defining feature of living systems. Most animals demonstrate adaptability by adjusting their position when confronted with external stimuli, a crucial skill for survival. How the brain dynamically regulates behavior in response to a changing environment remains a fundamental question in neuroscience. We investigated it using Danionella cerebrum, the smallest known vertebrate amenable to brain-wide functional imaging at cellular resolution across all developmental stages. We developed a 2D virtual reality system in which head-restrained larvae can navigate their visual environment. The system uses fluid dynamics estimates of the fish’s intended movements to restore their expected visual feedback. We observed that they can continuously stabilize their position when subjected to external visual flows of varying speed and direction. We mathematically modeled this regulation process with a system of delay differential equations that can exhibit limit cycle oscillations, consistent with the observed speed fluctuations. Moreover, we were able to perform calcium imaging during these virtual reality experiments and identify neural populations spanning the entire brain with activities that correlate with specific features of both behavior and visual stimulation. Notably, we found assemblies of neurons that activate differentially during spontaneous or visually-evoked swimming. In conclusion, our study not only significantly advances our understanding of how animals integrate sensory input in real-time to drive motor actions, but also introduces analysis and modeling tools with broader applicability, which may prove useful to other researchers in the field.