Gaussian Splatting as a Unified Representation for Autonomy in Unstructured Environments

--In this work, we argue that Gaussian splatting is a suitable unified representation for autonomous robot navigation in large-scale unstructured outdoor environments. Such environments require representations that can capture complex structures while remaining computationally tractable for real-time navigation. We demonstrate that the dense geometric and photometric information provided by a Gaussian splatting representation is useful for navigation in unstructured environments. Additionally, semantic information can be embedded in the Gaussian map to enable large-scale task-driven navigation. From the lessons learned through our experiments, we highlight several challenges and opportunities arising from the use of such a representation for robot autonomy. In environments such as those in Figure 1, traditional approaches often struggle to capture the complexity and variability of the scene, presenting challenges for autonomous navigation under such conditions. These capabilities are crucial for applications such as precision agriculture [1], forestry [2], search-and-rescue [3] and infrastructure inspection [4]. To address this, we present Gaussian splatting as a versatile representation for large-scale autonomy in unstructured outdoor environments.