Hello, I'm Tzu-Lin.
With over seven years of expertise in autonomous robotics, I am passionate about integrating cutting-edge technologies and innovative approaches. My mission is to bridge the gap between research and practical applications, continuously pushing the boundaries of what's possible in the field of autonomous systems.
OptiVerse Technology Ltd.
Taipei, Taiwan
Key Projects
People Counting System
- LiDAR-based real-time people tracking with resource management
- Intelligent LLM-powered analytics dashboard featuring real-time visualization, historical trends, and predictive occupancy patterns
Perimeter Security System
- Multi-sensor fusion for surveillance and real-time monitoring
- Advanced threat detection system using LiDAR-Camera fusion and skeleton analysis
Independent Consultant
Dec 2021 - March 2024 · 2 yrs 4 mos
Key Projects
Industrial Technology Research Institute
- Forklift Path Planning (AMR)
- 3D Localization and Mapping (AMR)
Foxconn Technology Group
- LiDAR Camera AR Navigation (Automotive)
Taipei Metro
- Intrusion Detection System (Railway)
M.Sc. in Electrical Engineering
Double Degree Program, Innovation & Entrepreneurship · 2 yrs 6 mos
Aalto University, Technische Universität Berlin
Contract Work
Unikie Oy
Mar 2020 - Dec 2021 · 1 yr 10 mos
Helsinki, Finland
Autonomous Valet Parking Development
- Developed real-time 3D CNN for LiDAR semantic segmentation
- Implemented 3D LiDAR-camera fusion for object detection
- Developed path planning with collision avoidance for autonomous parking
- Partially contributed to IAA Mobility 2021, Munich
Academic Projects
Technische Universität Berlin
Nov 2020 - Mar 2021 · 5 mos
Berlin, Germany
Robotics Project
- Developed solutions for robotic grasping using PCA and autoencoder analysis
- Worked with ContactPose Dataset to analyze human grasping behavior
- Created simplified representations of human grasp for robotic hand applications
Aalto University
Oct 2019 - Dec 2019 · 3 mos
Espoo, Finland
Research Project in Machine Learning and Data Science
- Modified BerkeleyAutomation/dex-net for parallel gripper grasp generation
- Utilized triangulation mesh completion for object surface reconstruction
- Implemented script to generate parallel grasps on different provided objects
Industrial Technology Research Institute (ITRI)
Jun 2017 - Jul 2019 · 2 yrs 2 mos
Hsinchu, Taiwan
Contract Work
- Acceleration design with CUDA for lidar point cloud classification and clustering
- Neural network acceleration using TensorRT for multiple camera streams
- Worked with NVIDIA Drive Platform Series and NVIDIA Jetson Platform Series