Skip to content
View Zhengsw03's full-sized avatar
6 followers · 7 following
  • FeisiLab
  • Changsha

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Block or report Zhengsw03

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Zhengsw03/README.md

S.W. Zheng | Robotics, robot learning and autonomous flight

Robot Learning · Autonomous Flight · Perception & Control
FeisiLab · Changsha, China

Repositories · Robot Learning · Email

About

My interests span robot learning, motion planning, and control. I work on practical robotics pipelines, from simulation workspaces to learning and deployment on real hardware.

Honors

  • RMUA National First Prize
  • RM AWARD Candidate
  • RM AWARD Algorithm Nomination Award

Robot Learning

RL-Token-Pi05-open

An end-to-end RL Token pipeline for the SO-101 arm: demonstration collection, π0.5 fine-tuning, critical-phase detection, online reinforcement learning on real hardware, and deployment.

SO-101 task comparison: VLA baseline on the left and RL Token policy on the right
VLA baseline (left) / RL Token policy (right) · SO-101 real-hardware demo

Python · PyTorch · LeRobot · Vision-Language-Action · Online RL

Aerial Robotics

RflyArena

A UAV controller testing and deployment framework spanning SITL simulation, HITL validation, and real-world flight.

Circle (left) · Figure-8 (right)

NMPC

RflyArena NMPC circle trajectory demo RflyArena NMPC figure-8 trajectory demo

PID/SO(3)

RflyArena PID/SO(3) circle trajectory demo RflyArena PID/SO(3) figure-8 trajectory demo

RL

RflyArena RL circle trajectory demo RflyArena RL figure-8 trajectory demo

Workspaces

Project Focus
Rflysim_fastlio_ws FAST-LIO and LiDAR localization workspace for RflySim.
C++ · ROS · Livox · State Estimation
Rflysim_fuel_ws UAV planning and simulation workspace with LiDAR and depth-camera launch workflows.
ROS · RflySim · Planning · Simulation

Toolkit

Python, C++, PyTorch, ROS, Linux and Git

Area Interests
Robot Learning VLA models · Reinforcement learning · Robot manipulation
Planning & Control Motion planning · MPC / NMPC · Trajectory tracking
Perception & Systems LiDAR localization · ROS / ROS 2 · PX4 · Simulation
Reading & Exploration

These upstream open-source projects are part of my reading and learning, not my original work.

Perceive. Plan. Learn. Act.
More experiments · siweizheng03@163.com

Pinned Loading

  1. Rflysim_fastlio_ws Rflysim_fastlio_ws Public archive

    C++ 1

  2. Rflysim_fuel_ws Rflysim_fuel_ws Public archive

    Makefile 1

  3. RflySim/RflyArena RflySim/RflyArena Public

    Python 1

  4. RL-Token-Pi05-open RL-Token-Pi05-open Public

    RL Token pipeline for the SO-101 arm: π0.5 fine-tuning, RL Token training, critical-phase classifier, online RL and deployment.

    Python 46