C's Robotics Paper Notes @RoboReading

Recording daily paper notes by @chongzitazhang who is interested in robot learning and control, with special focus on legged robots. ///I'm picky/// Joined March 2024

Tweets

424
Followers

3K
Following

1
Likes

9

C's Robotics Paper Notes @RoboReading

a day ago

First Order Model-Based RL through Decoupled Backpropagation (DMO) machines-in-motion.github.io/DMO/ Simulation rollouts, learned model for first order optimization

0 11 81 4K 46

Download Image

C's Robotics Paper Notes @RoboReading

a day ago

arxiv.org/abs/2509.02530 Manipulation as in Simulation: Enabling Accurate Geometry Perception in Robots Train a neural depth image filter to bridge sim real gap

0 6 38 2K 32

C's Robotics Paper Notes @RoboReading

a day ago

Attention based RL controller for generalized legged locomotion

Science Robotics @SciRobotics

5 days ago

Attention based RL controller for generalized legged locomotion

5 9 66 16K 30

Download Gif

1 0 20 2K 8

C's Robotics Paper Notes @RoboReading

a day ago

RL + GNN for tight space multi arm goal reaching

Science Robotics @SciRobotics

2 days ago

RL + GNN for tight space multi arm goal reaching

2 10 88 24K 28

Download Image

0 0 28 2K 14

C's Robotics Paper Notes @RoboReading

a week ago

arxiv.org/pdf/2508.21043 HITTER: A HumanoId Table TEnnis Robot via Hierarchical Planning and Learning Model based planning (with mocap perception) + RL stylistic tracking

1 26 160 9K 74

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

Aimbot: A Simple Auxiliary Visual Cue to Enhance Spatial Awareness of Visuomotor Policies arxiv.org/pdf/2508.08113 Drawing a visual aim makes visual policies perform better.

0 1 27 2K 19

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

MolmoAct: Action Reasoning Models that can Reason in Space arxiv.org/pdf/2508.07917 VLM/VLA with manipulation things and visual reasoning capabilities

1 15 139 10K 83

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

ODYSSEY: Open-World Quadrupeds Exploration and Manipulation for Long-Horizon Tasks arxiv.org/abs/2508.08240 Combined VLM-based global planning, action library, map-based navigation, local VLM-based planning, RL-based loco-mani WBC, and put the things into a large sim benchmark

0 1 23 2K 20

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion arxiv.org/abs/2508.08241 train tracking policies -> distill with diffusion -> use custom scores for downstream tasks without retraining. It's the followup of their previous SIGGRAPH work at RAI

0 22 132 7K 63

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

Uncertainty-aware Accurate Elevation Modeling for Off-road Navigation via Neural Processes arxiv.org/pdf/2508.03890 using Neural Processes for large-scale elevation mapping

0 1 16 1K 6

Download Image

C's Robotics Paper Notes @RoboReading

4 weeks ago

Interesting overview and discussion on dex hands arxiv.org/abs/2508.05415 Do Robots Really Need Anthropomorphic Hands?

0 0 17 1K 7

C's Robotics Paper Notes @RoboReading

a month ago

Confidence Calibration in Vision-Language-Action Models arxiv.org/abs/2507.17383

0 2 24 916 9

Download Image

C's Robotics Paper Notes @RoboReading

a month ago

From Seeing to Experiencing: Scaling Navigation Foundation Models with Reinforcement Learning metadriverse.github.io/s2e/ RL finetuning improves pretrained vision navigation

0 3 19 4K 9

Download Image

C's Robotics Paper Notes @RoboReading

2 months ago

jiaweihe.com/dexvlg DexVLG: Dexterous Vision-Language-Grasp Model at Scale Object segmentation -> optimized grasp per segment and VLM generated caption -> big dataset to train VLG model

1 22 110 6K 61

Download Image

C's Robotics Paper Notes @RoboReading

2 months ago

Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections compliant-residual-dagger.github.io human correction as the expert signal in dagger to finetune the policy, supported by compliant hardware design.