Joel Andrew Miller

Machine Learning Engineer | Computer Vision Researcher | Signal Processing Specialist

View Featured Project Explore All Research Let's Connect Download Resume
Joel Miller Illustration

About Me

I am a Machine Learning Engineer and Computer Vision Researcher specializing in signal processing and data-driven solutions. Currently working as an Aerospace Software Engineer at the International Test Pilots School (ITPS) Canada, I develop ML-powered systems for aviation applications, including computer vision algorithms for helmet-mounted displays and signal processing pipelines for real-time sensor data analysis.

My research focuses on applying machine learning techniques to physiological signal analysis, computer vision for autonomous systems, and statistical signal processing. I have published work on driver behavior analysis using heart rate signals and have extensive experience building ML pipelines for sensor fusion, image processing, and time-series analysis. My expertise spans deep learning, classical ML algorithms, and signal processing methods applied to real-world problems in transportation and aerospace domains.

Core Technical Skills

  • Machine Learning
  • Computer Vision
  • Signal Processing
  • Deep Learning
  • Python
  • TensorFlow/PyTorch
  • Statistical Analysis
  • Sensor Fusion
  • Time-Series Analysis
  • Data Science

Experience

February 2025 - Present

Aerospace Software Engineer

International Test Pilots School - London, Ontario

  • Leading R&D efforts to develop a helmet-mounted display system using computer vision algorithms for real-time information overlay and signal processing techniques for sensor data fusion
  • Implementing machine learning models for pilot state detection and adaptive display systems
  • Developing signal processing pipelines for multi-modal sensor data (IMU, GPS, video) integration
September 2023 - December 2023

R&D Engineering Intern

National Research Council Canada - London, Ontario

  • Developed computer vision and signal processing pipelines to transform multi-modal sensor data (LiDAR point clouds, camera images, GPS signals) into simulated environments for autonomous vehicle testing
  • Implemented data preprocessing algorithms for sensor fusion and coordinate transformation using MATLAB and Python
  • Built ML-based data validation and quality assessment tools for sensor data streams
May 2023 - December 2025

M.E.Sc Software Engineering

Western University - London, Ontario

  • Research focus: Machine learning and signal processing for autonomous vehicle systems
  • Developing ML models for driver state detection using physiological signal analysis (heart rate, EEG, eye-tracking)
  • Applying statistical signal processing methods (E-Tests, Poisson modeling) to analyze driver behavior patterns
  • Published first-author research on signal processing applications in transportation (Springer Nature, 2025)
  • Part of the AiX Lab (AI & Autonomous Systems) and EMRC Lab (Engineering & Machine Learning Research)
May 2021 - September 2022

Software Engineer / DevOps Intern

IBM - Markham, Ontario

  • Developed automated testing frameworks for IBM's Order Management System (OMS)
  • Built frontend components and Dockerized virtual machines for test case execution and monitoring
  • Mentored new interns and worked with Angular, Docker
September 2018 - April 2023

B.E.Sc Software Engineering with IBM Coop

Western University - London, Ontario

  • Completed my undergraduate degree in Software Engineering.

Featured Publication

Navigating the Handover

IEEE Review | August 2024

Published in IEEE Open Journal of Vehicular Technology

Tags: First Author | Review Paper | Machine Learning | Autonomous Vehicles | Human-AI Interaction

A comprehensive review of takeover requests in Level 3 autonomous vehicles, analyzing human-centered design approaches and exploring the Operational Design Domain (ODD) concept for safe autonomous operation.

  • First-authored comprehensive review paper
  • Published in prestigious IEEE journal
  • Identified key research gaps in AV handovers
Read Paper

At the Heart of Intersections: Analyzing Their Influence on Driver Heart Behaviour

Springer Nature | 2025

Published in Data Science for Transportation

Tags: First Author | Signal Processing | Statistical Analysis | Machine Learning | Data Science | Transportation

This paper investigates the physiological impact of intersections on driver heart rate, addressing a gap in prior research. Using video and heart rate data from the Honda Research Institute, we apply E-Tests to analyze Poisson distributions of HR events occurring within and outside intersections. Our findings support the hypothesis that intersections significantly influence driver heart rate, indicating heightened stress or cognitive load at these critical road junctures. This research has implications for urban traffic design and vehicle technology development in automated driving environments.

  • First-authored research paper investigating driver physiology at intersections
  • Utilized statistical analysis (E-Tests) on real-world driving data from Honda Research Institute
  • Found significant evidence that intersections influence driver heart rate and stress levels
  • Open-source code available for reproducibility and further research
Read Paper View Code

Research Projects

AiX Lab Simulation Environment

Research Infrastructure | 2024

Tags: Computer Vision | Signal Processing | Data Collection | Machine Learning

Simulation Setup with Triple Monitors

A comprehensive simulation environment for autonomous driving research with integrated computer vision and signal processing pipelines. Features multi-modal sensor data collection (video, physiological signals, vehicle dynamics) and real-time processing capabilities for ML model training and validation.

  • Built data collection infrastructure for ML model training with synchronized multi-modal sensors
  • Developed signal processing pipelines for physiological and vehicle dynamics data
  • Created computer vision tools for driver behavior analysis and gaze tracking
  • Integrated triple-monitor setup with comprehensive data logging for ML datasets

Reinforcement Learning for Autonomous Driving

Machine Learning Research | 2023

Tags: Deep Learning | Reinforcement Learning | Computer Vision | Autonomous Vehicles

CARLA Simulator Environment

Developed a deep reinforcement learning agent for autonomous navigation using computer vision-based perception. Implemented DQN and PPO algorithms with CNN feature extractors for processing camera inputs, trained on complex driving scenarios in the CARLA simulator, achieving robust navigation capabilities in diverse environments.

  • Implemented deep RL agents (DQN, PPO) with CNN-based computer vision processing
  • Designed reward functions and state representations for autonomous driving tasks
  • Trained models on complex multi-agent scenarios with varying weather and traffic conditions
  • Integrated with CARLA simulator using OpenAI Gym interface for reproducible ML experiments

Machine Learning Demo: MNIST Digit Recognition

Interactive Deep Learning Demo - Draw a single digit (0-9) below

This demonstrates a convolutional neural network (CNN) trained on the MNIST dataset for handwritten digit classification. The model processes your drawing using computer vision preprocessing techniques.

CNN Prediction

-

This CNN model demonstrates computer vision and deep learning capabilities. The model uses image preprocessing, feature extraction, and classification. Try different drawing styles to see how the model performs! 😄

Get in Touch

Email: joelmiller0430@gmail.com

Phone: (905) 933-7920

LinkedIn: LinkedIn Profile

ORCID: ORCID Profile

Resume: Download Resume (PDF)