864 papers
This paper proposes a two-stage reward curriculum that decouples task-specific objectives from behavioral terms to improve exploration and training stability in robotic reinforcement learning, demonstrating superior performance and robustness across multiple environments compared to direct full-reward training.
The paper proposes SeedPolicy, a self-evolving diffusion policy enhanced by a novel Self-Evolving Gated Attention (SEGA) module that efficiently compresses long-horizon observations, enabling state-of-the-art performance in robotic manipulation tasks with significantly fewer parameters than existing vision-language-action models.
This paper proposes a complexity-aware adaptive inference framework for Vision-Language-Action models that dynamically routes execution to "Act," "Think," or "Abstain" based on task complexity, leveraging a vision-only detector to optimize resource allocation and prevent failures while achieving high accuracy with minimal training data.
This paper proposes SkillsCrafter, a novel framework for lifelong language-conditioned robotic manipulation that mitigates catastrophic forgetting and enhances generalization by employing Manipulation Skills Adaptation to retain and inherit knowledge, and Skills Specialization Aggregation to leverage common semantic subspaces for skill similarity and aggregation.
This paper introduces "Critic in the Loop," a tri-system framework that dynamically coordinates a high-level Vision-Language Model for global reasoning and a fast Vision-Language-Action model for reactive execution via a lightweight visual critic, thereby achieving robust, state-of-the-art performance in long-horizon robotic manipulation by balancing semantic depth with real-time control.
This paper presents a digital twin-driven dual-arm robotic system that integrates RGBD sensing, tactile feedback, and state-of-the-art Visual Language Models to achieve robust, real-time textile classification and foreign object detection for automated sustainable recycling.
This paper advocates for expanding the role of collaborative robots in physical rehabilitation beyond repetitive motion training to assist therapists and patients throughout the entire therapy process, thereby addressing access barriers while highlighting key challenges in safety, user-state understanding, and workflow integration.
This paper introduces Curve-induced Dynamical Systems on Smooth Manifolds (CDSM), a real-time framework that generates stable and adaptable robotic behaviors on Riemannian manifolds and Lie groups by constructing dynamical systems with tangential and normal components relative to a nominal curve, demonstrating superior accuracy and efficiency in both benchmarks and practical robotic applications.
This paper presents a Vehicle-in-the-Loop and digital twin-based framework that integrates a physical test vehicle on a dynamometer with a synchronized virtual environment to enable safe, cost-effective, and realistic validation of autonomous driving algorithms on centralized E/E architectures without requiring individual ECU testing or intermediate software layers.
The paper proposes HD-ExpIt, a framework that iteratively refines hierarchical diffusion policies for language-conditioned manipulation by leveraging environment feedback to autonomously discover and distill successful behaviors, thereby aligning the planner with the controller's capabilities and achieving state-of-the-art performance on the CALVIN benchmark.
The paper proposes Latent Policy Steering (LPS), a robust offline reinforcement learning method that achieves state-of-the-art performance by using a differentiable one-step MeanFlow policy to backpropagate original-action-space Q-gradients directly to a latent actor, thereby eliminating the need for proxy latent critics and sensitive hyperparameter tuning while ensuring policies remain within dataset support.
This paper presents a geometrically exact, constraint-free static model for continuum parallel robots that utilizes kinematic embedding and a fourth-order Magnus approximation to solve nonlinear equilibrium equations on a product manifold, with experimental validation confirming its accuracy under large deformations and external loads.
The paper introduces UltraDexGrasp, a framework that generates a large-scale synthetic dataset of 20 million bimanual grasping trajectories to train a point-cloud-based policy capable of achieving robust zero-shot sim-to-real transfer with an 81.2% success rate on diverse real-world objects.
This paper presents a unified CT-to-simulation pipeline that integrates patient-specific cochlear reconstruction with a differentiable Cosserat-rod model and contact-aware robotic planning to minimize intracochlear trauma and prevent insertion failures during robotic cochlear implantation.
This paper presents Omni-Manip, an end-to-end LiDAR-driven visuomotor policy that leverages a Time-Aware Attention Pooling mechanism to process 360° panoramic point clouds, enabling humanoid robots to perform robust dexterous manipulation in large, cluttered workspaces without the need for frequent repositioning or reliance on narrow-field-of-view RGB-D cameras.
OpenFrontier is a training-free, lightweight navigation framework that achieves robust zero-shot generalization in open-world environments by leveraging vision-language models to identify semantic frontiers as visual anchors for goal-directed navigation, eliminating the need for dense 3D mapping, policy training, or model fine-tuning.
This paper introduces MPPI-DK, a computationally efficient model predictive path integral control framework that leverages a learned linear deep Koopman operator to replace nonlinear dynamics for faster trajectory sampling, achieving near-optimal performance with significantly reduced computational costs in both simulation and real-world robotic applications.
This paper introduces CliReg, a novel deterministic loop closure validation algorithm that replaces RANSAC with a maximal clique search on feature compatibility graphs to achieve more robust and accurate 3D LiDAR-based SLAM performance under noisy and ambiguous conditions.
ROScopter is a modular, researcher-focused multirotor autopilot built on ROSflight 2.0 and ROS 2 that accelerates simulation and hardware testing while achieving performance comparable to state-of-the-art systems with a significantly reduced codebase.
This paper introduces PhysiFlow, a physics-aware, multi-brain Vision-Language-Action framework that leverages latent flow matching and robust tracking to enable efficient, stable, and semantically guided whole-body control for humanoid robots.