I am an undergraduate Robotics Engineering student at Beijing Institute of Technology, working at the intersection of robotics, autonomous driving, embodied AI, 3D scene representation, and AI systems engineering.
I enjoy turning research prototypes into real systems—from perception, state estimation, and navigation to inference deployment, communication architecture, testing, and fault recovery. Beyond getting a demo to run, I care about reliability, reproducibility, and graceful recovery when things go wrong.
- Currently building — 3DGS visual localization, instance navigation, and editable road-scene generation
- Exploring — ROS 2, Nav2, Habitat, VLA/VLN, world models, and AI agents
- Engineering — reliable real-robot systems, AI inference deployment, autonomous-driving infrastructure, and open-source software
- Let's talk about — embodied AI, 3D scene intelligence, robotics middleware, AI infrastructure, and research automation
- Highlights — 20 merged upstream PRs; Kaggle Bronze Medal (solo, top 9.9%)
|
A batch-wise extension of Point-LIO featuring within-window motion undistortion, batched EKF updates, and parallel computation for high-bandwidth LiDAR-inertial odometry.
|
A 3D annotation and AI-assisted pre-labeling platform for Formula Student Driverless point clouds, with quality gates, OpenPCDet export, and a TensorRT inference backend.
|
|
An end-to-end mobile manipulation system integrating base navigation, object detection, 3D grasp-point generation, and arm control—from environmental perception to autonomous grasping.
|
MuJoCo simulation and control experiments for RoboMaster manipulator tasks, exploring visual pose estimation, forward and inverse kinematics, and autonomous grasping.
|
|
Research notes on 3DGS, world models, and editable 3D scenes for instance navigation, visual localization, and autonomous-driving scene generation.
|
A collection of Unitree G1/Go2, RoboMaster, and Formula Student Driverless projects spanning perception, state estimation, navigation, communication, inference deployment, and hardware integration.
|
Contributed 20 merged pull requests across AI infrastructure, robotics middleware, developer tooling, and scientific computing projects.
| Project | Pull Request | Focus |
|---|---|---|
| Hugging Face TRL | #6348 | DPO / KTO file handling |
| OpenRLHF | #1261 | Qwen3.5 and ZeRO-3 compatibility |
| vLLM | #48153 | Mistral Large 3 weight loading |
| NVIDIA CCCL | #9785 | Compiler compatibility |
| CycloneDDS | #2425 | Signal-mask handling |
| ROS 2 CLI | #1257 | interface show --no-comments |
| ros2_control | #3454 | update_rate boundary handling |
| OpenCV | #29487 | Calibration documentation |
View all 20 merged PRs
- huggingface/trl #6348
- OpenRLHF/OpenRLHF #1261
- vllm-project/vllm #48153
- opencv/opencv #29487
- NVIDIA/cccl #9785
- ccxt/ccxt #29192
- rust-lang/cargo #17203
- rsheftel/pandas_market_calendars #469
- Unclecheng-li/poc-lab #20
- eclipse-cyclonedds/cyclonedds #2425
- duckdb/duckdb #23773
- Farama-Foundation/Gymnasium #1618
- eclipse-iceoryx/iceoryx2 #1818
- Unclecheng-li/poc-lab #21
- vyperlang/vyper #5185
- ros2/ros2cli #1257
- ros-controls/ros2_control #3454
- nautechsystems/nautilus_trader #4443
- Velocidex/velociraptor #4921
- biomejs/biome #10915
|
|
