Sampling-based Kinodynamic Planning with World Models
This paper revisits kinodynamic sampling-based motion planning for world models, using the learned model as the propagator of a tree search in latent space.
ProofPaper ↗
Key points
- Visual world models predict the consequences of actions from images, but planning with them is typically done by optimizers such as CEM and MPPI.
- Our key insight is that the sampling strategies devised for high-dimensional search in configuration space apply directly to the latent space of a world model, where they rapidly expand through intermediate states that myopic optimizers never sample.
- We present two exploration strategies for the high-dimensional latent space, nearest-neighbor-guided RRT and projection-based KPIECE, with GPU-native batched expansion for efficient device utilization.
- We evaluate with an off-the-shelf DINO-WM on four simulated domains and a real-world block-pushing task.
Sources (1)
- [1]Sampling-based Kinodynamic Planning with World ModelsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 03:01 AM
This paper revisits kinodynamic sampling-based motion planning for world models, using the learned model as the propagator of a tree search in latent space.
Visual world models predict the consequences of actions from images, but planning with them is typically done by optimizers such as CEM and MPPI.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 9, 2026Latent World Action Model for GUI Agents
- Oct 9, 2026RSI-Anything: Unlocking Recursive Self-Improvement with World Models
- Oct 8, 2026Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction
- Oct 8, 2026DreamTrue: Action-Faithful Robot World Model with Counterfactual Post-Training
- Oct 8, 2026Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks
- Oct 8, 2026OuroWorld: Bringing Any 3D World Alive as Diverse, Endlessly Looping 3D Cinemagraphs