archive
Every paper Pith has read. Search by title, abstract, or pith.
2900 papers in cs.RO · page 9
-
Retrained VMAF cuts error in teleoperation video ratings by 27 percent
Beyond VMAF: Towards Application-Specific Metrics for Teleoperation Video
-
Refinement cuts UAV localization error in cold starts and against faults
Uncertainty-Aware 3D Position Refinement for Multi-UAV Systems
-
Shared tokens unify bimanual robot control
CUBic: Coordinated Unified Bimanual Perception and Control Framework
-
Flow-adaptive MMD preserves ergodic coverage in dynamic domains
Asymptotically Optimal Ergodic Coverage on Generalized Motion Fields
-
Two demonstrations yield 90% success under OOD robot conditions
SID: Sliding into Distribution for Robust Few-Demonstration Manipulation
-
Rotational latents lift VLA success to 98% on LIBERO
RotVLA: Rotational Latent Action for Vision-Language-Action Model
-
Trajectory augmentation enables offline RL from limited suboptimal data
Trajectory-Level Data Augmentation for Offline Reinforcement Learning
-
BlockVLA adapts pretrained autoregressive vision-language-action models into discrete…
BlockVLA: Accelerating Autoregressive VLA via Block Diffusion Finetuning
-
This paper compares three robot interaction styles—passive (no help)
Exploring Human-Robot Collaboration: Analysis of Interaction Modalities in Challenging Tasks
-
Stereo vision and location priors boost real-world robot navigation
What Limits Vision-and-Language Navigation ?
-
Robot success in crowds rises 14% with human intention forecasts
HCSG: Human-Centric Semantic-Geometric Reasoning for Vision-Language Navigation
-
Decoupled planes enable linear scaling in large VLA RL
D-VLA: A High-Concurrency Distributed Asynchronous Reinforcement Learning Framework for Vision-Language-Action Models
-
D-VLA delivers linear speedup for trillion-param VLA models
D-VLA: A High-Concurrency Distributed Asynchronous Reinforcement Learning Framework for Vision-Language-Action Models
-
Galilean symmetry keeps INS consistent with unknown delays
Galilean State Estimation for Inertial Navigation Systems with Unknown Time Delay
-
Ranking sensor readings localizes gas sources without calibration
Calibration-Free Gas Source Localization with Mobile Robots: Source Term Estimation Based on Concentration Measurement Ranking
-
This paper builds a unified computational model that treats an athlete's rigid skeleton…
Dynamics Computation of Soft-Rigid Hybrid-Link System and Its Application to Motion Analysis of an Athlete Wearing Sport Prosthesis
-
EvObj segments 3D objects by evolving synthetic priors unsupervised
EvObj: Learning Evolving Object-centric Representations for 3D Instance Segmentation without Scene Supervision
-
The paper introduces ERPPO, a MAPPO variant that trains a DSA learner to estimate object…
ERPPO: Entropy Regularization-based Proximal Policy Optimization
-
Movable circle cuts collision overestimation at single-circle speed
MoCCA: A Movable Circle Probability of Collision Approximation
-
Depth-aware path planning achieves uniform seafloor coverage over 99%
Multi-Depth Uniform Coverage Path Planning for Unmanned Surface Vehicle Surveying
-
High-level planner plus specialized tools lifts long-horizon robot success
Towards Long-horizon Embodied Agents with Tool-Aligned Vision-Language-Action Models
-
Language contacts lift dexterous grasp success to 98 percent
SECOND-Grasp: Semantic Contact-guided Dexterous Grasping
-
Paired visuals teach RL which changes to ignore in robot tasks
What to Ignore, What to React: Visually Robust RL Fine-Tuning of VLA Models
-
Kinematic tangent cone separates non-transversal branches
Identification of Non-Transversal Bifurcations of Linkages
-
Truss robot manipulates objects with hybrid position-force control
Object Manipulation of the Variable Topology Truss system
-
Tactile pressure maps predicted from egocentric video
TouchAnything: A Dataset and Framework for Bimanual Tactile Estimation from Egocentric Video
-
A simple episode-wise relative frame for proprioceptive encoding delivers better…
When Absolute State Fails: Evaluating Proprioceptive Encodings for Robust Manipulation
-
Single policy unifies skills for wheeled-legged robots
MUJICA: Multi-skill Unified Joint Integration of Control Architecture for Wheeled-Legged Robots
-
Dual IMUs enable global exponential relative pose and velocity estimates
Relative Pose-Velocity Estimation Using Dual IMU Measurements and Relative Position Sensing
-
Semantic images calibrate robot dynamics uncertainty with guarantees
Local Conformal Calibration of Dynamics Uncertainty from Semantic Images
-
Swarms form sub-goal chains to transport objects past obstacles
Occlusion-Based Object Transportation Around Obstacles With a Swarm of Miniature Robots
-
Safety certification reduces to one-dimensional switching-time search
Distributionally Robust Safety Under Arbitrary Uncertainties: A Safety Filtering Approach
-
This paper develops a safety filtering method for nonlinear control systems that…
Distributionally Robust Safety Under Arbitrary Uncertainties: A Safety Filtering Approach
-
Directed factor keeps robot estimates clean during joint planning
DynoJEPP: Joint Estimation, Prediction and Planning in Dynamic Environments
-
Bipolar pneumatic controller cuts pressure errors by 12-36%
BiPneu: Design and Control of a Bipolar-Pressure Pneumatic System for Soft Robots
-
Physics-informed network reconstructs tube robot shapes under 1% error
Few-Shot Physics-Informed Neural Network for Shape Reconstruction of Concentric-Tube Robots
-
Enhanced EWC cuts forgetting in vision-language models by 78%
Lifelong Learning in Vision-Language Models: Enhanced EWC with Cross-Modal Knowledge Retention
-
This paper tested how 100 people perceived emotions from short videos of Reachy Mini
Emotional Expression in Low-Degrees-of-Freedom Robots: Assessing Perception with Reachy Mini
-
Adaptive smoothing unlocks non-convex Pareto policies in robotics
Adaptive Smooth Tchebycheff Attention for Multi-Objective Policy Optimization
-
Autonomy stack works across drone and robot types
The Unified Autonomy Stack: Toward a Blueprint for Generalizable Robot Autonomy
-
Five-layer MLOps setup lets ADS fleets learn from shared data
A Five-Layer MLOps Architecture for Connected Automated Driving
-
Residual risk now accounts for localization uncertainty in AVs
Belief-Space Residual Risk for Automated Driving under Localization Uncertainty
-
Hybrid RL-DWA boosts microrobot 3D navigation and deformation
3D RL-DWA: A Hybrid Reinforcement Learning and Dynamic Window Approach for Goal-Directed Local Navigation in Multi-DoF Robots
-
Search method reveals VLM failure modes in driving and robotics
Revealing Interpretable Failure Modes of VLMs
-
Joint optimization finds better robot locomotion than separate design
COSMIC: Concurrent Optimization of Structure, Material, and Integrated Control for robotic systems
-
Mixing keeps hybrid filters from losing contact branches
Support-Safe Variational Hybrid Filtering for Contact-Mode and Sparse-Law Recovery
-
Structured learning predicts uncertainty for risk-aware off-road MPC
Multistep Belief Space Dynamics Learning For Risk-Aware Control
-
Intent labels let driving policies beat human scores
Driving Intents Amplify Planning-Oriented Reinforcement Learning
-
Intent conditioning lets driving RL exceed human demonstrations
Driving Intents Amplify Planning-Oriented Reinforcement Learning
-
Unified VLA first to beat humans on long-tail driving benchmark
MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Driving