arXiv:2606. 03756v1 Announce Type: cross Abstract: We introduce Neural Navigation Functions (Neural-NF), a learned reactive navigation function capable of zero-shot transfer across unseen environment geometries.
arXiv:2607. 20743v1 Announce Type: cross Abstract: Trajectory planning is a fundamental problem in robotics, requiring the generation of collision-free and efficient trajectories in a potentially complex environment.
arXiv:2607. 17574v1 Announce Type: cross Abstract: Reinforcement-learning navigation policies for legged robots select actions reactively from current observations and short-term memory, with limited capacity to anticipate how moving obstacles will evolve in the near future.
arXiv:2606. 01098v1 Announce Type: cross Abstract: Generative action policies based on diffusion or flow matching excel in behavior cloning, yet their iterative sampling is prohibitive for high-frequency robot control.
arXiv:2604. 12474v3 Announce Type: replace-cross Abstract: In many robotic tasks, agents must traverse a sequence of spatial regions to complete a mission.
arXiv:2602. 05031v2 Announce Type: replace Abstract: Planning with a learned model remains a key challenge in model-based reinforcement learning (RL).
arXiv:2603. 08862v2 Announce Type: replace-cross Abstract: Autonomous navigation in highly constrained environments remains challenging for mobile robots.
arXiv:2606. 18828v1 Announce Type: cross Abstract: Traditional approaches place intelligence in the agent, whether as a learned policy or a search procedure.
arXiv:2605. 04568v3 Announce Type: replace-cross Abstract: State-of-the-art model-based Reinforcement Learning (RL) approaches either use gradient-free, population-based methods for planning, learned policy networks, or a combination of policy networks and planning.
arXiv:2606. 14585v1 Announce Type: cross Abstract: Generative dynamics models enable planning in challenging robotic systems, but safe deployment requires reliably detecting policy-induced out-of-distribution (OOD) transitions.
Neural world models coupled with model predictive control (MPC) replan at every environment step to bound accumulated prediction error, but this incurs substantial computational overhead. Reusing a cached plan reduces this overhead, yet its effectiveness depends on how prediction mismatch propagates through the local dynamics.
arXiv:2606. 09311v1 Announce Type: new Abstract: Joint Embedding Predictive Architectures (JEPAs) have shown promising world modeling capabilities, enabling planning in latent space by optimizing action trajectories using methods like the Cross-Entropy Method (CEM).
Conventional visual navigation policies often struggle with myopic decision-making and mode collapse in complex environments. While world models offer a promising alternative, existing paradigms typically isolate perception, generation, and control, failing to capture their shared spatio-temporal dynamics.