Robotics
Research on robot control, manipulation, navigation, and human-robot interaction.
cs.RO · 524 papersSearch-based Robustness Testing of Laptop Refurbishing Robotic Software
PROBE is a search-based method for robustly testing object detection models in laptop refurbishing robots, significantly outperforming random search.
Multi-Robot Coordination in V2X Environments
This paper proposes a V2X framework with new services (RAS, RMCS) for decentralized multi-robot coordination, integrating non-V2X users in urban traffic.
Cross-Modal Navigation with Multi-Agent Reinforcement Learning
CRONA is a Multi-Agent Reinforcement Learning framework for cross-modal navigation, improving collaboration via auxiliary beliefs and a centralized critic.
ReActor: Reinforcement Learning for Physics-Aware Motion Retargeting
ReActor uses bilevel optimization and RL to retarget human motion onto robots, producing physically plausible and robust motions for imitation learning.
Lie Group Formulation of Recursive Dynamics Algorithms of Higher Order for Floating-Base Robots
This paper presents Lie-group formulations for higher-order dynamics of floating-base robots, showing quadratic scaling in computational cost.
OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulation
OA-WAM introduces an object-addressable world action model that decomposes scenes into persistent object slots for robust robot manipulation under scene shifts.
GA3T: A Ground-Aerial Terrain Traversability Dataset for Heterogeneous Robot Teams in Unstructured Environments
GA3T is a new real-world dataset for heterogeneous air-ground robot teams, enabling collaborative perception in diverse unstructured environments.
TouchDrive: Electronics-Free Tactile Sensing Interface for Assistive Grasping
TouchDrive enables precise assistive robotic grasping using an electronics-free tactile sensing interface with pneumatic feedback.
Reconstruction or Semantics? What Makes a Latent Space Useful for Robotic World Models
This paper evaluates reconstruction vs. semantic latent spaces for robotic world models, finding semantic spaces better for policy-relevant tasks.
AssistDLO: Assistive Teleoperation for Deformable Linear Object Manipulation
AssistDLO is an assistive teleoperation framework for manipulating deformable linear objects, combining state estimation, visual assistance, and geometry-aware shared autonomy.
Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation
VISER is a new visually realistic benchmark for robot manipulation, bridging the sim-to-real gap with high-fidelity assets and strong real-world correlation.
CKT-WAM: Parameter-Efficient Context Knowledge Transfer Between World Action Models
CKT-WAM enables parameter-efficient knowledge transfer between World Action Models by injecting teacher context into student text embeddings.
Structure-Preserving Gaussian Processes Via Discrete Euler-Lagrange Equations
Lagrangian Gaussian Processes (LGPs) learn physically consistent dynamics from sparse position data, ensuring stable long-term predictions by preserving geometric structure.
RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI
RobotEQ introduces the first benchmark for active intelligence, assessing if embodied AI can understand and adhere to social norms without explicit commands.
Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
ProCompNav is a two-stage framework that uses comparative judgment and binary questions to efficiently navigate ambiguous user queries.
When to Trust Imagination: Adaptive Action Execution for World Action Models
This paper introduces an adaptive execution method for World Action Models (WAMs) that verifies future predictions against reality, improving robotic manipulation efficiency and robustness.
EA-WM: Event-Aware Generative World Model with Structured Kinematic-to-Visual Action Fields
EA-WM is a generative world model that uses structured kinematic-to-visual action fields to improve robot interaction dynamics and geometry in generated videos.
VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts
VLA-GSE introduces a novel parameter-efficient fine-tuning framework using generalized and specialized experts to boost VLA model adaptation for robotic control tasks.
CredibleDFGO: Differentiable Factor Graph Optimization with Credibility Supervision
CredibleDFGO improves urban GNSS positioning by explicitly training for reliable covariance estimates using differentiable factor graph optimization and proper scoring rules.
Monitoring autonomous persistent surveillance missions using invariance
This paper introduces a compositional runtime monitor for autonomous persistent surveillance, enabling efficient monitoring of black-box robot systems in large environments.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.