Daxiao Robotics and NTU S-Lab Release World Model Mundo

Daxiao Robotics recently partnered with Nanyang Technological University's S-Lab to release Mundo, a world model built for specific environments that can be explored and interacted with. Mundo supports scene reconstruction, free-viewpoint exploration, and spatiotemporal simulation capabilities similar to those of the Atlas multimodal world model. Simply put, it can use a small number of photos or videos to build a three-dimensional scene that can be freely walked through and viewed from different angles, and it can also let robots practice navigation and manipulation within it. This capability provides support for robots entering specific commercial scenarios. In Daxiao Robotics' world model technology system centered on Kairos, Mundo supplements environment modeling, training data generation, and simulation training capabilities, helping robots complete scene adaptation and policy optimization. Building a dedicated world model from a small number of images, supporting long-path generation across spaces Mundo expands a small number of images or videos into a three-dimensional space that can be continuously explored; the same set of reference images can support multiple camera paths; and for larger scenes, it can dynamically select context and continuously generate exploration footage ranging from tens of seconds to the minute level. Specifically, Mundo starts from contextual imagery of real locations, first builds a three-dimensional proxy to determine the spatial relationships between different views, and then generates new viewpoints along a specified camera path. In the scenes demonstrated this time, inputting 1 to 12 images is enough to explore around the furniture and spatial structure in a room. The same set of reference images can serve multiple paths, allowing the observation position and angle to change freely. During the modeling process, a three-dimensional detector provides initial positions and extents for the main objects, and a vision-language model agent further adjusts the proxy scene, through rendering, comparison, and correction, so that the scene structure corresponds to the input imagery. For example, an office can be explored from beside a desk around to a window, and a residence can be entered from a courtyard into the interior. When the camera moves, the input images constrain the layout, appearance, and details that have already been captured, while the model infers and completes the areas that have not been captured. Adding reference views provides more real evidence for the scene and reduces the uncertainty of completion. This capability can also start from a single generated image or painting, extending its structure and visual style to new viewpoints. For larger environments, Mundo uses dozens of contextual images to support long-path generation ranging from tens of seconds to the minute level. Objects have physical properties, and pushing, pulling, and collisions all have computable responses For a robot to practice in a scene, the environment must be able to respond to actions. A drawer being pulled open, a chair being pushed, or an object falling to the ground all require corresponding position changes and contact relationships. Mundo can edit the world through code, describing an object's name, size, and pose through a scene program, and configuring physical properties such as mass and friction for it. The script defines actions and applied forces, and the physics engine computes the objects' responses under gravity and contact constraints, yielding motion trajectories, contact positions, and collision forces. Mundo then combines the environment reference images to render these changes into the corresponding scene. This gives the environment explicit operational entry points. Users can drag trash cans, push monitors, pull chairs, or open drawers, and can also change object materials and gravity conditions to observe responses under different settings. States such as velocities, contact points, and impact forces during the physical process can also be displayed in sync with the generated frames. In this process, the 3D proxy and scene program are responsible for recording "where objects are and what changes have occurred," the physics engine is responsible for computing motion, and the video model is responsible for presenting visual results consistent with the environment. Geometry, physics, and appearance each assume clear responsibilities and together form an interactive environment. Making every interaction have traceable states and computable processes is the key to moving scenes from viewable to usable. Mundo introduces physical interaction into reconstructed environments, providing a concrete basis for robots to practice actions and analyze results. Connecting navigation, manipulation, and four-dimensional simulation in the same environment to build a robot training ground In the same customized scene, Mundo further supports robot movement and manipulation, connecting environment reconstruction to robot testing and training. For navigation, the 3D proxy provides information such as object categories, positions, obstacles, and traversable areas, on which basis the path planner finds routes. For manipulation, Mundo establishes geometric and physical properties for the objects involved in interaction, allowing robots of different forms to attempt the same task, such as putting an apple into a bowl, placing a can onto a tray, or stacking blocks. Mundo also extends spatial exploration into the temporal dimension. For dynamic events recorded by multiple cameras, the process can be presented from new viewpoints; for simulated events built from images, motion is computed through gravity and contact dynamics. The industrial value of a world model must be demonstrated by its contribution to real tasks. In the world model technology system of Dexmal Robotics with Kairos at its core, Mundo supplements specific environment modeling and simulation training capabilities, connecting on-site collection, data generation, and policy optimization, and serving robots as they enter real work scenarios.