[Experiment Prep] Importing a Custom Robot into MuJoCo
Import a custom robot model into MuJoCo to lay the groundwork for building simulation environments for future reinforcement learning.
Blog
Notes on robotics, vision, and systems.
[Note-Taking Tools] Markdown Syntax and the Obsidian Editor
An introduction to Markdown, the powerful note-taking syntax I have long used, and Obsidian, an editor based on Markdown syntax.
[Paper Notes] Residual Skill Policies: Learning an Adaptable Skill-based Action Space for Reinforcement Learning for Robotics
A skill-unit-based reinforcement learning method that decomposes tasks into skills, establishes mappings between skills and states, and learns a skill-selection policy
[Paper Notes] DEX: Demonstration-Guided Reinforcement Learning with Efficient Exploration for Task Automation of Surgical Robot
A surgical robot reinforcement learning method based on an improved DDPG+BC approach, developed by Yun-Hui Liu's team at The Chinese University of Hong Kong
Reinforcement Learning and Imitation Learning Bug List (Updated Occasionally)
A record of bugs that may occur while setting up reinforcement learning environments and running projects, along with their solutions
[Paper Notes] Robot Motion Imitation with Reinforcement Learning
Survey and notes on reinforcement learning-based robot motion imitation methods.
[Paper Reproduction] MimicPlay: Learning Robot Skills from Human Demonstrations
A long-horizon imitation learning method from Fei-Fei Li's team at Stanford University that learns by watching human actions.
Installing and Using the PCL Library
The most widely used point cloud library in C++
Quickly Implementing Three-Dimensional Reconstruction with NeRF (instant-ngp)
Instant-NGP, short for Instant Neural Graphics Primitives, uses multiresolution hash encoding to solve the efficiency problem encountered when parameterizing NeRF with a fully connected neural network. It greatly accelerates network training, reducing three-dimensional reconstruction time from several hours to several seconds.
【Motion Imitation】Robot Motion Planning Methods Learned from Human Demonstrations
A survey of research on methods for robot motion generation and planning learned from videos of humans grasping objects.