695 papers
Seed2Scale is a self-evolving data engine that overcomes data bottlenecks in embodied AI by synergizing a lightweight "SuperTiny" model for robust data collection with a large Vision-Language Model for autonomous quality verification, enabling a target model to achieve a 131.2% performance improvement starting from just four seed demonstrations.
SAIL is a test-time scaling framework that enhances one-shot robot imitation learning by reframing trajectory generation as an iterative refinement process guided by Monte Carlo Tree Search, an automated retrieval archive, and a vision-language model-based scoring mechanism, thereby significantly improving success rates across diverse manipulation tasks.
This paper demonstrates that explicitly reducing observation dimensionality and implementing locality-aware credit assignment in a communication-free multi-agent system enhances robustness and performance in asymmetric 3D pursuit-evasion tasks within cluttered environments.
EndoSERV is a novel vision-based navigation system for endoluminal robots that overcomes challenges like tissue deformation and label scarcity by combining segment-to-structure odometry with real-to-virtual transfer learning to achieve accurate localization without requiring real-world pose labels.
This paper introduces HIMoS, a hierarchical multi-modal planning framework that enables Autonomous Underwater Vehicles to efficiently search for and sample sparse benthic targets like coral colonies at a fixed altitude by integrating a global topological route optimizer with a local differentiable belief propagation planner, thereby outperforming traditional exhaustive and adaptive sampling strategies in high-fidelity simulations.
PhaForce is a phase-scheduled visuomotor policy that enhances contact-rich manipulation by coordinating a slow, vision-dominant diffusion planner with a fast, force-driven corrector to enable high-frequency, phase-aware residual corrections, achieving an 86% success rate and superior adaptability compared to existing baselines.
This paper presents a communication-free multi-UAV coordination framework that leverages onboard anisotropic 3D LiDAR for simultaneous SLAM, obstacle detection, and neighbor tracking, enabling safe and scalable navigation in complex, GNSS-denied environments like dense forests.
MoMaStage is a structured vision-language framework that enables robust long-horizon indoor mobile manipulation by guiding task planning through a topology-aware Skill-State Graph and ensuring execution reliability via a closed-loop mechanism that triggers semantic replanning upon detecting physical deviations, all without requiring explicit scene mapping.
The paper proposes StructBiHOI, a hierarchical framework that combines a jointVAE for long-term planning, a maniVAE for frame-level refinement, and a Mamba-based diffusion denoiser to achieve stable, physically plausible, and semantically aligned long-horizon bimanual hand-object interaction generation.
This paper identifies value-scale amplification as the primary cause of instability in non-linear value decomposition for offline multi-agent reinforcement learning and proposes a scale-invariant value normalization technique to stabilize training, ultimately providing a practical recipe to unlock the full potential of offline MARL.
This paper proposes an AI-driven perception method with hierarchical human intention prediction to enable mobile robot chemists in self-driving laboratories to proactively distinguish between human preparatory actions and transient interactions, thereby overcoming the inefficiencies of passive obstruction detection and streamlining human-robot coordination in shared-access scenarios.
This paper proposes the AFOP-ML framework, an automatic feature optimization-enabled prototypical network that achieves rapid few-shot tactile recognition of both shapes and materials with high accuracy and robustness against perturbations, effectively addressing the challenges of data scarcity and time-consuming training in robotic applications.
The FoMo dataset presents a comprehensive, multi-season collection of over 64 km of diverse robot navigation data from a boreal forest, featuring significant environmental changes like heavy snow and vegetation growth to challenge and evaluate the robustness of state-of-the-art odometry and SLAM systems.
This paper presents an adaptive entropy-driven sensor selection policy within a camera-LiDAR particle filter that dynamically switches between modalities to optimize tracking accuracy and continuity for single-vessel surveillance, validated through real-world maritime deployment.
The paper proposes R2F, an LLM-free framework for zero-shot open-vocabulary object navigation that repurposes ray frontiers as direction-conditioned semantic hypotheses to achieve competitive performance with real-time execution, eliminating the latency and computational overhead of iterative large-model queries.
LAR-MoE is a two-stage framework that decouples unsupervised skill discovery from policy learning by regularizing expert routing to align with a learned latent representation, enabling robots to achieve high success rates in heterogeneous manipulation tasks without requiring manual skill annotations.
The paper proposes STRIDE, a hybrid dynamics learning framework that combines a Lagrangian Neural Network for energy-consistent rigid-body mechanics with Conditional Flow Matching for stochastic residual interaction forces, achieving significant improvements in long-horizon prediction and contact force accuracy for robotic systems in unstructured environments.
3PoinTr is a novel method that pretrains robust robot manipulation policies from casual, unconstrained human videos by using a transformer architecture to predict 3D point tracks as an embodiment-agnostic intermediate representation, enabling sample-efficient learning with minimal robot demonstrations.
This paper introduces an open-source, robot-agnostic surgical robotics platform featuring a deterministic, closed-form RCM controller and full-stack ROS integration, which achieves sub-millimeter precision and expert-level trajectory smoothness in autonomous laparoscopic tasks across phantom, ex vivo, and in vivo porcine models.
This paper introduces ProReFF, a feature field model that learns relative object co-occurrence distributions from unlabeled observations to guide robotic search agents, achieving 20% higher efficiency than strong baselines and up to 80% of human performance in the Matterport3D simulator.