Tuesday, November 17, 2026
12 papers added this day. New papers stay for 14 days.
Tactile sensing 1
Dexterous manipulation 1
Grasping 1
Imitation learning and data collection 1
Robot foundation models 1
RL and sim-to-real 1
Legged locomotion 1
Humanoids and mobile manipulation 2
Computer vision for robots 1
Human motion prediction 1
Exoskeletons and wearable robots 1
ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters
November 2026
Attach a frozen 552B mixture-of-experts VLM to a trainable action DiT and compare it with 2B-32B backbones under identical training. The 552B policy learns LIBERO-Goal from 10 demonstrations better than a 2B backbone does from 50 (87.8% vs 85.6%) and inserts a charger on a real bimanual robot in 9/10 trials vs 4/10, at 216 ms per action chunk.
Multi-CONDOR: Learning Multistable Motion Primitives from Demonstration
November 2026
Learns one stable motion policy from demonstrations that converges to any of several goals or to a continuous goal contour. On 2D benchmarks no start state ends away from a goal (StableNODE leaves 0.31% and 7.29%), and trajectory error is about 72% lower than per-goal models picked by a state machine.
SG-ODOM: Event-Driven Proprioceptive State Estimation with Stability-Gated Contact Updates for Legged Robots
November 2026
SG-ODOM is leg odometry that applies kinematic corrections only after a foot's contact has settled, plus timeout updates that cap IMU-only stretches at about 0.1 s. Its smoother version has the lowest average drift on ANYmal, Spot and a Unitree G1, 0.157 m against 0.208 m for ANYbotics' onboard estimator on seven GrandTour missions.
ARMFUL: Learning Whole-Arm Grasp Configurations for Single-Arm Pickup of Large Objects
November 2026
A simulated dataset of 1.7 million whole-arm grasps on 793 large objects, each stress-tested by shaking and pushing, trains a diffusion model that chooses how one Franka arm wraps around a stool, box or rack to lift it. The pipeline lifts 85.4% of training and 69.9% of held-out objects in simulation, and 70 of 100 real trials on 10 IKEA objects.
DAPPT: Domain-Aware Policies from Procedural Trajectories for Sim-to-Real Dexterous Tool Use
November 2026
Train a dexterous hand-arm policy in rigid-body-only simulation to track procedurally generated tool trajectories, with the trajectory parameters as a policy input that can be tuned per task. Picking the best parameters raises task metrics 1.5-3.8x in a particle simulator, and the policy transfers zero-shot to a real robot on six tasks such as sand scooping and play-dough rolling.
Cleanly-DINO: Efficiently Learning Visual Task Spaces for Robotics
November 2026
Fits a small projection on frozen DINOv2 features from a few cleaning videos, so that every image patch gets a dirty-to-clean progress score. It separates clean from dirty patches with 0.763 AUROC on held-out videos. A real robot that wipes the least clean patch first was rated at about 83% of human cleaning quality.
SGIP: Learning Scene-Grounded Interaction Priors for Scene-Aware Human Motion Prediction
November 2026
SGIP first predicts which object or open region of a 3D scene a person is heading for, then forecasts their trajectory and full-body pose toward it, using an LLM-guided teacher only at training time. Without gaze or intent at inference, it cuts GIMO trajectory error to 549 mm from 642 mm for the best comparable baseline.
Towards Probabilistically Safe Reinforcement Learning: Theory and Deployment on Humanoid and Aerial Robots
November 2026
Extend a safe-RL method with training-time safety guarantees (RSGF-RL) so it accepts PPO's biased gradient estimates and general trajectory-level constraints. Policies trained with it transfer zero-shot to a Unitree G1 that squats, crawls and walks within height and attitude limits that vanilla PPO breaks, and to a Crazyflie drone that avoids moving obstacles with a 0.1 m safety margin.
Continuous Lower-Limb Joint-Angle Estimation from sEMG and IMU: The Conditional Value of Cross-Activity and Cross-Subject Data under Update-Matched Controls
November 2026
An offline study that estimates hip, knee and ankle angles from leg sEMG and an IMU, and measures which extra training data actually help a new user against a control matched in training updates. Adding the user's own other activities raises R² from 0.722 to 0.788, of which +0.048 is information rather than extra training.
Ego2Sim2Humanoid: Agentic Real-to-Sim for Learning Loco-Manipulation VLAs without Robot Demonstrations
November 2026
E2S2H turns recordings from egocentric smart glasses into physically checked whole-body training data for a Unitree G1 humanoid, then fine-tunes the GR00T N1.7 VLA on that data alone. Without any real-robot demonstrations, the policy lifts a bottle in 9 of 20 real trials and puts on a hat in 5 of 20.
Leaning Over to Reach Far-Away Objects
November 2026
A sampling-based MPC planning 20 MuJoCo rollouts at 34 Hz on a laptop lets a Unitree H1-2 humanoid brace its forearm on a table and reach about 50 cm beyond its unbraced workspace. On hardware it reaches the commanded target in 89% of 54 trials and completes the full reach-and-recovery cycle in 81%.
TMA: Sim-to-Real Forceful Humanoid Manipulation via Tactile-Driven Motor Adaptation
November 2026
A humanoid learns to open doors with round knobs entirely in simulation, adapting its grip online from 2 s of fingertip force history. On a Unitree G1 with four real knobs it opens the door in 27 of 40 trials, against 6 for a reactive tactile policy and none for RMA.
Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.
