Henry Kou

Henry Kou

(he/him)

MS Robotics Student and Research Associate

Carnegie Mellon University Robotics Institute

Biorobotics Lab

About Me

I am an MS Robotics student at Carnegie Mellon University with a background in electrical and computer engineering and a focus on state estimation, control theory, motion planning, and embedded robotic systems. As a research associate in the Biorobotics Lab, I work on building reliable, sensor-driven robots that connect theory with hardware and real-world autonomy.

Education

MS Robotics

2024-08-01
2026-05-31

Carnegie Mellon University

BS Electrical Engineering

2017-09-01
2021-06-30

University of California, Los Angeles

Interests

Robotics Embedded systems Sensors Control theory Motion planning Optimization

Evidence-based skill map

Skills & Expertise

Evidence-backed robotics, software, and hardware strengths drawn from the portfolio, resume, and author profile.

View full matrix

Robotics Subsystems

Motion planning and search

Weighted A*, time-expanded pursuit, RRT, RRT-Connect, RRT*, PRM, Dijkstra-style graph search, and STRIPS-style task planning.

Control and optimization

Model Predictive Path Integral control, sampling-based MPC, whole-body locomotion costs, process-control timing, PID, and linear feedback coursework.

Programming Languages

Python

Simulation, planning analysis, visualization, manipulation assignments, MPPI work, and local developer tooling.

C and C++

Search/planning implementations, embedded robot work, and robotics coursework projects.

Tools and Frameworks

MuJoCo

Wheeled quadruped simulation tasks and controller evaluation.

PyBullet and MEngines

Panda manipulation simulation, collision checking, point-cloud capture, and grasp attempts.

Hardware and Platforms

Unitree Go2W

Wheeled quadruped model used for whole-body MPPI locomotion experiments.

Panda manipulator simulation

IK, pick-and-place, grasp search, and force-closure coursework.

Robotics demo gallery

Robotics Demos

Project-linked videos and local demo media for robotics platforms, simulation, and planning work.

View all demos

EigenBot Modular Limb Demo

Modular robot limb experiments connecting force sensing, full-limb behavior, and early neural-controller work on repeated robot modules.

modular robotics force sensing neural control

Medusa Space Arms Demo

Carriage-mounted space-arm testbed work for coordinated manipulation, sensing, and control on a physical robotics platform.

space robotics manipulation controls

MPPI Wheeled Quadruped Simulation

Whole-body MPPI controller for the Unitree Go2W in MuJoCo, with wheel-aware actions and rough-terrain simulation tasks.

MPPI wheeled quadruped MuJoCo
Featured Projects
ARPA-E Pipe Inspection featured image

ARPA-E Pipe Inspection

Confined-space pipe inspection robotics work with crawler hardware, mapping, and misalignment visualization.

Boeing Material Deposition featured image

Boeing Material Deposition

In-progress process-control research for ultra-large-format additive manufacturing that adapts deposition timing to uncompensated vibrations.

EigenBot featured image

EigenBot

Modular robot work involving force sensing, limb experiments, and preliminary neural controller results.

Medusa Space Arms featured image

Medusa Space Arms

Space-arm robotics work using a carriage-mounted arm testbed for coordinated motion, sensing, and manipulation research.

A* Pursuit of Dynamic Target

A C++ robotics planner that uses weighted A* in a time-expanded grid to intercept a moving target while replanning from the robot's current state.

Manipulation Course Assignments

Python manipulation coursework in MEngines covering rigid-body transforms, forward kinematics, screw coordinates, contact mechanics, RRT-Connect planning, antipodal grasp search, …

MPPI Wheeled Quadruped

Whole-body MPPI controller for the Unitree Go2W wheeled quadruped, extending legged locomotion control with wheel-torque actions, wheel-aware costs, and MuJoCo simulation tasks.

Queue up for Claude

A local, usage-aware job queue for Claude Code that stages coding tasks, injects per-project agent context, and runs queued work near the end of a Claude.ai usage window.

Sampling Based Planners

C++ sampling-based motion planners for a high-DOF planar robot arm, comparing RRT, RRT-Connect, RRT*, and PRM on collision-free joint-space planning tasks.

Symbolic Planner

A C++ STRIPS-style symbolic planner for grounding action schemas and searching over Blocks World, Blocks Triangle, and robot-fire-extinguisher planning domains.