Mohammad Rouful Azim
Electronic Engineering student at Hochschule Hamm-Lippstadt with a foundation in Autonomy Technologies from FAU Erlangen-Nürnberg. My work focuses on building transparent engineering simulations, noise-resistant state estimators (EKF), and DSP-driven sensor pipelines.
I enjoy working across the hardware-software boundary testing digital filters, validating sensor discrepancies, and designing robust data pipelines in Python and C. I prefer rigorous, measurable experimentation over complex black-box abstractions.
Direct Contact: rimon.rouful11@gmail.com
Featured Engineering Systems
Live Telemetry & ControlsMotor Condition Monitoring & Fault Classification
Built a simulated motor condition-monitoring system that uses current, vibration and temperature signals to identify abnormal operating conditions. The pipeline applies digital signal processing and frequency-domain feature extraction before classifying operating modes with a Random Forest model. A separate rule-based protection layer handles deterministic overcurrent and overtemperature conditions.
Wheel-Slip Detection & Adaptive EKF for Differential-Drive Robots
Built a state-estimation pipeline for a differential-drive robot operating under wheel-slip conditions. The system combines wheel-encoder odometry with IMU measurements using an Extended Kalman Filter (EKF). A Random Forest classifier detects slip from encoder–IMU discrepancies and dynamically adjusts the measurement covariance to reduce pose-estimation errors during traction loss.
Device-Free Human Activity Detection Using Wi-Fi CSI
Explored whether Wi-Fi Channel State Information (CSI) can be used to detect human movement without cameras or wearable devices. The system processes CSI amplitude variations using sliding-window statistics and classifies changes in activity to identify movement and prolonged inactivity. The prototype is designed around device-free sensing, making it suitable for privacy-sensitive indoor environments.
Work & Research Experience
2 Engineering RolesStudent Researcher (HiWi) / Chair of Sensor Systems
2025 — PresentSupport research into state estimation for autonomous indoor mobile platforms. Benchmark Extended Kalman Filter (EKF) variants against high-noise odometry slip, and maintain the department's ROS 2 simulation stack used across multiple thesis test benches.
Technical Team Lead, Perception / Autonomy Technologies
2024 — 2025Led a 6-person perception group implementing localization for an autonomous delivery platform. Handled sensor calibration pipelines across optical wheel encoders, MEMS IMUs, and 2D planar LiDAR. Cut trajectory drift by tuning dynamic covariance thresholds against physical ground-truth markers.
Stack & Practical Frequency
Deterministic state machines, HAL design, hardware timers, and interlock protection logic.
DSP feature engineering, FFT spectral decomposition, and machine learning classifiers.
Engineering Discipline · Asymmetric Safety Architecture
Modern machine monitoring often tries to make machine learning responsible for safety shutdowns. In high-power industrial electrical drives, that approach violates functional safety (IEC 61508). Probabilistic classifiers belong strictly in an advisory role for early mechanical wear. Hard safety limits (overcurrent, thermal runaway, sensor saturation) must remain deterministic in C-level logic, executing in deterministic interrupt contexts without heap allocation.