Systems Engineering · Software for Mechatronics · Reinforcement Learning

Moving towards dependable, human-centred autonomous systems.

First-Class Master of Engineering (International): Cardiff University with a KAIST exchange. I’m an aspiring mechatronics engineer focused on physics-based modelling & simulation with hardware-integrated ROS deployments.

  • ROS (Noetic)
  • Python
  • C++
  • RTAB-Map
  • PPO
  • State-space & digital control
  • Sustainability
  • Continuous Improvement
  • Teamwork & Communication

The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System

2024-09-01—2025-05-01 ROS Noetic · Python · RTAB-Map · EEG logging (Emotiv, CYKIT)
The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System — image 1
The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System — image 2
The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System — image 3
The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System — image 4
The Design and Evaluation of Haptic Feedback Methods and Environmental Impacts on a Wearable Navigation System — image 5

Built a wearable navigation system (RGB-D VSLAM, APF navigation, haptic actuation) with EEG event logging. Trials showed fixed reference haptic cues outperformed rotational haptic ques on success, safety, and reaction time.

Reinforcement Learning-Based Control of a Quadruped Agent in a Simulated Sumo Arena

2023-08-01—2024-01-01 PyTorch · RaiSim/RaisimGym · CUDA · NumPy · TensorBoard
Reinforcement Learning-Based Control of a Quadruped Agent in a Simulated Sumo Arena — image 1
Reinforcement Learning-Based Control of a Quadruped Agent in a Simulated Sumo Arena — image 2
Reinforcement Learning-Based Control of a Quadruped Agent in a Simulated Sumo Arena — image 3
Reinforcement Learning-Based Control of a Quadruped Agent in a Simulated Sumo Arena — image 4

Trained full ANYmal with joint-torque control and a two-stage curriculum; analysed learning curves, hyperparameters, and emergent tactics under adversarial contact.

View all projects