Khuzema Habib
Robotics Engineer · Full-Stack Builder · Researcher
I'm a Robotics Engineer with a Master of Engineering in Robotics from the University of Maryland, College Park, and a background in mechanical engineering. I work full-stack across the robotics stack — mechanical design and fabrication, sensor and actuator integration on edge computers, ROS 2 autonomy, perception and VLAs, and bringing those systems onto real hardware in the field. That ranges from Nav2 and EKF localization on custom tracked platforms to teleoperation pipelines, tactile-augmented manipulation, and adaptive quadrotor control. I care most about closing the loop from simulation and research ideas to reliable, deployed robots.
Research Interests
Skills
Programming Languages
Robotics & AI
Hardware
Mechanical & CFD
Education
Master of Engineering in Robotics
University of Maryland, College Park · 2024 – 2026
GPA 3.67/4
Bachelor of Technology in Mechanical Engineering
Manipal Institute of Technology · 2019 – 2023
GPA 8.03/10
Work Experience
- Built autonomous navigation for a custom tracked platform (two differential-drive units on passive pivots): derived the kinematic model and tuned an EKF fusing IMU, magnetometer, wheel odometry, and RTK-GPS (private NTRIP over Tailscale).
- Implemented a Nav2 stack with STVL costmaps from a ZED X Mini and a behavior tree for field coverage, tight-furrow driving, and automated row-to-row U-turns.
- Field-tested inference for Pi0.6/0.7 VLA policies on a bimanual strawberry-picking robot, tuning inference settings (action horizon, time warp, chunk size) and debugging failures to report per-configuration success rates.
- Automated the annotation pipeline (YOLO pre-labels in CVAT, VLM filtering of false detections), cutting annotation time 65% and raising classifier accuracy up to 72% [from 61% baseline].
- Integrated sensors and actuators on edge computers over CAN, SPI, I2C, UART, and TCP/IP with ROS2.
- Setup a teleoperation/data collection pipeline with support for DAgger on a UR3e arm for imitation learning models using an affordance-guided framework that distills a minimal set of semantic 2D keypoints from a text prompt and a single image.
- Built Stanford Doggo (quadruped), added carbon fiber rods for improved chassis stiffness and assembly.
- Conducted flight trials to collect trajectory datasets for Crazyflie drone control systems, developing Contextual NeuroMHE Controllers to improve adaptability and reduce trajectory deviations in turbulent environments.
- Autonomous indoor navigation (SLAM/LiDAR) and mobile manipulation with Jetson Nano on Turtlebot2 platform and a LeRobot-based arm to execute RL policies through LLM prompts.
- Integrated GelSight tactile data with Pi0 to augment VLA policies for manipulation on a UR5e with a custom UMI gripper.
- Guided students in implementing perception and planning pipelines on the DJI Mobile Manipulator platform.
- Mentored 40+ students on robot control fundamentals, including kinematics, dynamics, and trajectory generation.
- Conducted weekly office hours and lab sessions, providing technical support and clarifying complex concepts in robot control.
- Assisting 35+ students in understanding control system design (PID & LQR).
- Covering real-world applications in robotics for system modeling and optimization.
- Conducted ANSYS simulations to analyze piezo-electric materials for energy harvesting.
- Mathematical Modeling of Flow Environments leveraging CFD with Ansys Fluent.
- Acoustics and Vibration Analysis for Acoustic Black-Hole applications.
- Performed 3D Simulations to analyze the efficacy and flow patterns for the various desuperheaters in the Fisher Product Catalog.
- Assisted the Project Management team at Emerson Dubai for the Fisher Control Valves Business Unit which primarily caters to Oil and Gas companies in the MENA region.
- Assisted with Shop Floor Inspection of valves and components as well as Customer Support on Specification sheets, coordinating with various Subject Matter Experts (SMEs) to resolve any issues.
- Performed 2D and 3D CFD simulations for a rocket thruster to identify optimal fuel and oxidizer injector configurations.
- Gained hands-on experience with simulation tools to improve injector design for propulsion efficiency.
Robots & Sensors
Robots
Sensors
Projects
Contact-aware Hierarchical Policy for Insertion
Visuo-tactile HDMI insertion with a skill-token-augmented ACT policy supervised by a VLM + heuristic auto-labeling pipeline.
- Used an Isaac Lab simulation and inference setup with a VLM + heuristic pipeline (Gemma 4) that auto-labels task phases (Skills) from RGB, tactile, and end-effector signals.
- Supervised a skill-token-augmented ACT policy for visuo-tactile HDMI insertion that raised success from 25% to 40% over the ACT baseline.
Autonomous Mobile Manipulator
Built and developed a fully autonomous robot capable of computer vision-based object localization and servo-controlled manipulation.
- Built and Assembled a Mobile Robot from Scratch with an integrated gripper for mobile manipulation tasks.
- Raspberry Pi 4B platform with a camera and ultrasonic sensor, along with an Arduino nano for IMU data collection.
- Integrated sensor-fused navigation with wheel odometry and IMU data.
- Pick-and-place tasks with subsystems integration.
- Successfully completed the final challenge of the competition, where the robot needed to autonomously find RGB blocks and place them in a target zone.
- Added a custom designed LED ring indicator to the robot for navigation state visualization.
Autonomous Vehicle Behaviour Planning In CARLA
Behavior-tree-based decision making and trajectory planning for lane change, following, and overtaking, with LiDAR-fusion SLAM for odometry on the KITTI dataset.
- Behavior Planning: Uses Behavior Trees to decide high-level maneuvers (LANE_KEEP, FOLLOW, LANE_CHANGE, STOP_FOR_LIGHT, RECOVERY).
- Trajectory Planning: Generates candidate trajectories in Frenet coordinates and selects the optimal path based on a weighted cost function.
- Vehicle Control: Implemented PID controller for longitudinal speed and Stanley controller for high-fidelity lateral steering.
- State Machines: Integrated tracking for in-progress lane changes and a robust recovery mode for off-road or stuck scenarios.
Stanford Doggo Inspired Quadruped
Ongoing
Building an agile and durable quadruped robot inspired by Stanford Doggo to use as a testbed for future projects.
- Fabrication (3D Printing and CNC machining) and Assembly
- Wiring and Odrive setup with custom firmware, Teensy 3.5 Microcontroller for control with Xbee for Wireless communication
- Chassis reinforcement with the use of Carbon Fiber Rods over Sheet Metal
- Testing and fine-tuning the build to ensure proper functionality — Need to fix some issues with weaker PLA components snapping under load when walking
- Future integration with ROS2 for Autonomous Navigation and easier integration with other subsystems
- Future improvements to mechanical design to add additional degree of freedom to the legs
A Star Path Planner for Turtlebot3
Developed an A* path planner for a Turtlebot3 robot to navigate through a maze-like environment. Validated on Gazebo Simulation as well as real-world testing.
- Developed a 2D map environment using half-planes and semi-algebraic equations with configurable robot clearances.
- Integrated a trajectory controller using proportional control to translate planned waypoints into linear and angular velocities.
- Validated path execution across multiple simulation platforms, including Gazebo, the Falcon Simulator, and Matplotlib.
- Processed real-time odometry data (/odom) to handle pose estimation and heading angles via quaternion transformation.
- Optimized movement by incorporating differential drive constraints, allowing user-defined RPM inputs for wheel velocities.
Pursuit RRT* Path Planner for Turtlebot3
Developed an RRT* path planner for a Turtlebot3 robot to navigate through a maze-like environment to catch a moving target.
- Developed a 2D RRT* path planner for a maze-like environment to catch a moving target.
- The planner implements a pursuit strategy to catch a randomly moving target by using RRT* to plan a path to the target's current position.
- Extended this to a Gazebo simulation with Turtlebot3 robots, where the pursuer robot uses RRT* to plan a path to catch the target robot.
- The pursuer robot extracts the current pose of the evader robot and plans a path, and is considered caught when the robot is in view and a certain distance away using a depth camera.
- The Evader robot implements proportional control to follow the waypoints, subscribing to /odom to extract pose and heading angles from the quaternion, then publishing linear and angular velocities to /cmd_vel.
Multi-Agent Reinforcement Learning for Drone Swarm Control
Custom MAPPO/IPPO implementation in PyTorch for coordination tasks in gym-pybullet-drones, achieving a 3× performance improvement over Stable-Baselines3.
- Developed custom Multi-Agent Proximal Policy Optimization (MAPPO) and IPPO algorithms from scratch using PyTorch for high-efficiency swarm control.
- Extended the gym-pybullet-drones environment to include complex multi-agent scenarios: hovering, spiral formation, and leader-follower navigation.
- Implemented a centralized critic in MAPPO to resolve global credit assignment, significantly reducing return variance during training.
- Achieved a 3x performance improvement over Stable-Baselines3 (SB3) in coupled tasks, reducing Spiral RMSE to 0.25m.
- Optimized safety metrics, reaching a low collision rate of 2.1% while maintaining maximum episode stability of 240 steps.
- Eliminated the overhead of standard RL libraries (Rllib/MARLlib) by building a streamlined, flexible training pipeline.
LQR Controller for Double Inverted Pendulum Cart Problem
Implemented an optimal control strategy for a complex nonlinear system involving a cart and two coupled pendulums.
- Derived the nonlinear equations of motion for a cart with two pendulums using Euler-Lagrange equations.
- Linearized the complex nonlinear system about a specific equilibrium point to create a state-space representation.
- Developed an optimal LQR controller by penalizing state error and actuator effort through tuned Q and R matrices.
- Certified the closed-loop stability of the system using Lyapunov's indirect method and eigenvalue analysis.
- Designed Luenberger observers for observable output vectors to estimate system states from limited measurements.
- Implemented a Linear Quadratic Gaussian (LQG) controller to maintain optimal performance in the presence of process and measurement noise.
Visual Servoing with MPC
Designed MPC visual servoing controller for Turtlebot3 navigation using RealSense and Aruco markers.
- Model Predictive Control: Integrated a visual servo-based MPC framework to steer nonholonomic wheeled robots.
- Polar Kinematics: Transformed robot kinematics into a polar coordinate frame to ensure smoother steering and avoid Cartesian singularities.
- Real-time Optimization: Utilized CVXPY and OSQP Solver to solve Quadratic Programming (QP) optimization problems efficiently.
- Visual Servoing: Used a depth camera for position-based visual servoing and real-time depth feedback.
- Constraint Enforcement: Enforced operational constraints such as actuator saturation and velocity limits directly within the control design (Q and R matrices).
- Gazebo Validation: Validated the framework in simulation, demonstrating superior precision and speed over traditional PD control.
Publications
- Created a lightweight CFD-informed trajectory planner generating wind-optimized paths in <10s using FluidX3D.
- Validated system on Crazyflie 2.1 hardware, achieving a 12.5–58.7% reduction in path divergence and 24.6% better stability (mean jerk) by proactively anticipating aerodynamic disturbances.
Built the UR3e experimental setup (calibration, teleoperation, data collection) for a VLM-to-STL-to-MPC manipulation framework that enforces spatial, temporal, and logical task constraints, reaching 80% safe success on LIBERO-Constrained vs. 14% for SmolVLA.
Developed an affordance-guided keypoint selection framework enabling lightweight, real-time robotic manipulation with 82% success on unseen objects.
Developed a reinforcement learning-based adaptive controller enabling robust quadrotor trajectory tracking across diverse environments with 20.3% trajectory error reduction on Crazyflie Drones.