archive
Every paper Pith has read. Search by title, abstract, or pith.
2900 papers in cs.RO · page 15
-
Benchmark shows robots can't yet follow social rules unprompted
RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI
-
Yes/no questions that split candidate pools raise navigation success
ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
-
Binary questions on candidate pools raise navigation success
ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
-
Binary questions split candidate pools to resolve vague navigation queries
ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
-
Verifier lets robots execute longer action chunks when predictions match reality
When to Trust Imagination: Adaptive Action Execution for World Action Models
-
Verifier lets robots trust predicted actions longer for speed
When to Trust Imagination: Adaptive Action Execution for World Action Models
-
Robot actions projected as visual fields boost world model accuracy
EA-WM: Event-Aware Generative World Model with Structured Kinematic-to-Visual Action Fields
-
Spectral split turns VLA fine-tuning into 2.5% parameter updates
VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts
-
VLA-GSE adapts robot models with 2.51% of parameters
VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts
-
End-to-end scoring of GNSS covariance yields credible urban fixes
CredibleDFGO: Differentiable Factor Graph Optimization with Credibility Supervision
-
Compositional invariants monitor black-box robot surveillance
Monitoring autonomous persistent surveillance missions using invariance
-
MPCC tracks ornithopter paths to 9 cm at 3 m/s
Accurate Trajectory Tracking with MPCC for Flapping-Wing MAVs
-
Flapping-wing MAV tracks paths within 9 cm at 3 m/s
Accurate Trajectory Tracking with MPCC for Flapping-Wing MAVs
-
Diffusion model completes partial maps for robot goal finding
Plug-and-Play Label Map Diffusion for Universal Goal-Oriented Navigation
-
Residual RL turns human hand motions into robot skills
DexSynRefine: Synthesizing and Refining Human-Object Interaction Motion for Physically Feasible Dexterous Robot Actions
-
Vehicle LiDAR converts into usable roadside training data
Generating Roadside LiDAR Datasets from Vehicle-Side Datasets via Novel View Synthesis
-
Pulsed-jet robot reaches 0.5 m/s with origami mantle and glide
Cycle-resolved Cephalopod-Inspired Pulsed-Jet Robot With High-Volume Expulsion and Drag-Reduced Gliding
-
Cephalopod robot hits 3.8 body lengths per second with origami mantle
Cycle-resolved Cephalopod-Inspired Pulsed-Jet Robot With High-Volume Expulsion and Drag-Reduced Gliding
-
This paper runs simulations to compare Incremental Nonlinear Dynamic Inversion (INDI)…
A Comparative Study of INDI and NDI with Nonlinear Disturbance Observer for Aerial Robotics
-
Model finds resource-safe strategies for robots under mixed uncertainty
Resource-Constrained Robotic Planning in the face of Mixed Uncertainty
-
Adapters correct geometry loss for accurate 3D human-object contacts
MaMi-HOI: Harmonizing Global Kinematics and Local Geometry for Human-Object Interaction Generation
-
Triadic relations let robots adapt to new scenes and objects
TriRelVLA: Triadic Relational Structure for Generalizable Embodied Manipulation
-
Angular momentum penalty produces pendular robot gaits
On the Emergence of Pendular Structure in Multi-Contact Locomotion
-
Generated images lift forest species mapping F1 scores by over 15 points
Leveraging Image Generators to Address Training Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping
-
Terminal CBF constraint in MPC enlarges reachable sets as horizon grows
Maximal Controlled Invariant-MPC: Enhancing Feasibility and Reducing Conservatism through Terminal CBF Constraint in Safety-Critical Control
-
Normalized advantages let RL pick best chunk size per state
Adaptive Q-Chunking for Offline-to-Online Reinforcement Learning
-
Event cameras meet cooperative perception latency in real vehicular VLC
Real-world Latency Analysis of Vehicular Visible Light Communication with Multiple LED Transmitters and an Event-Based Camera
-
H∞ controller keeps quadrotor tracking despite online parameter shifts
Robust $\mathcal{H}_\infty$ Controller Design For INDI-Controlled Quadrotor Using Online Parameter Identification
-
Gripper ToF sensors predict grasp stability before contact
Contact-Free Grasp Stability Prediction with In-Hand Time-of-Flight Sensors
-
Counterfactual simulations help robots pick creative tools
Creative Robot Tool Use by Counterfactual Reasoning
-
se(3) control keeps PDEs intact for any-length flexible arms
Modular Lie Algebraic PDE Control of Multibody Flexible Manipulators
-
Lie algebra cancels interaction forces in flexible robot chains
Modular Lie Algebraic PDE Control of Multibody Flexible Manipulators
-
Tension feed-forward gives UAVs passive recovery from cable cuts
Passive Fault Tolerance through Tension-to-Thrust Feed-Forward: Hybrid Input-to-State Stability for Decentralized Multi-UAV Slung-Load Transport under Abrupt Cable Severance
-
TA* finds tracking paths 23 times faster than standard A*
Track A*: Fast Visibility-Aware Trajectory Planning for Active Target Tracking
-
Query density recalibrates 3D detectors under data shifts
Query2Uncertainty: Robust Uncertainty Quantification and Calibration for 3D Object Detection under Distribution Shift
-
Dual-barrier CBF filter keeps robots safe while mapping unknown spaces
A Closed-Form Dual-Barrier CBF Safety Filter for Holonomic Robots on Incrementally Built Occupancy Grid Maps
-
Q-values pulled from behavior cloning let robots learn online in hours
When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning
-
Consistency modules lift robot VLA success 21% and speed 2.3x
ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulation
-
Line-guided RL teaches bicycle robot five stunts without demos
LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stunts
-
Line guidance trains bicycle robot to perform five stunts on command
LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stunts
-
Model forecasts driver cabin reactions from traffic context
Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout
-
Neural decoder turns coarse MPM into high-detail tactile surfaces
Reduced-order Neural Modeling with Differentiable Simulation for High-Detail Tactile Perception
-
Embodied AI privacy requires lifecycle architecture
Position: Embodied AI Requires a Privacy-Utility Trade-off
-
Modular RL lets robot swarms coordinate with limited memory
Modular Reinforcement Learning For Cooperative Swarms
-
Reservoir lifts nonlinear dynamics to stable linear models
Koopman Identification of Nonlinear Systems via Reservoir Liftings
-
Uncertainty metric refines AX=YB calibration for 67% better accuracy
Optimal Uncertainty-Aware Calibration for the AX=YB Problem
-
Direct radar SLAM outperforms priors on 300 km of real data
Dr-PoGO: Direct Radar Pose-Graph Optimization
-
Test-time training adapts visual foresight VLA to OOD shifts
Test-Time Training for Visual Foresight Vision-Language-Action Models
-
Data variety beats complex models for robot eye tracking
Gaze4HRI: Zero-shot Benchmarking Gaze Estimation Neural-Networks for Human-Robot Interaction
-
Elastic bands fold flat 3D prints into functional robots
3D Printing of Passively Actuated Self-Folding Robots with Integrated Functional Modules