RESEARCH INTERNSHIP • FLANDERS MAKE (2025)
Autonomous AI Palletizer
Deep Reinforcement Learning 3D parcel stacking in Unity ML-Agents (C#), deployed to an industrial robotic arm via ROS2 & C++.
Reinforcement Learning
Unity ML-Agents (C#)
ROS2 & C++ Runtime
3D Physics Simulation
Real Robot Arm Deployment
Flanders Make
Simulation & Robotic Control Video Demonstration
Live Recording
Technical Subsystems
Unity ML-Agents Training (C#)
Parallelized 3D RL policy training on randomized box dimensions and arrival sequences.
ROS2 & C++ Robotic Deployment
C++ inference node communicating over ROS2 topics for physical robot arm trajectory execution.
Stability & Density Reward Shaping
Multi-objective reward function for center-of-mass balance, volumetric density, and anti-tipping.
Sim-to-Real Calibration
Friction, parcel deformation, and placement tolerance compensation for real-world reliability.
Want to discuss robotics & reinforcement learning?
I would be excited to talk about ROS2 pipelines, deep RL simulation, and industrial automation.