Marine Engineer specializing in autonomous marine systems, ship maneuvering, and control engineering.
Focus areas: physics-based vessel models, Nonlinear MPC (NMPC), Deep Reinforcement Learning (DRL), and digital twins—grounded in real engine-room operations and maritime industry constraints.
This repository is organized as a set of project “modules” (folders), each documenting an engineering deliverable relevant to modern maritime operations:
- Yacht Autonomous Docking Project/ — Hybrid DRL + NMPC autonomy stack for safe docking in confined waters
- Digital Twin Process for Pump System/ — Digital-twin workflow applied to a pump system (documentation + artifacts)
- Constraint Programming/ — Shipbuilding production scheduling optimization (constraint programming case study)
- CNC Machining with error detection/ — Manufacturing quality / error detection work (process & reporting)
Folder: Yacht Autonomous Docking Project/
A full-stack autonomy pipeline for low-speed, high-risk maneuvers in marina conditions—designed to be constraint-aware, safe, and operationally interpretable.
Core architecture
- 3‑DOF physics-based maneuvering model (MMG) for realistic vessel response
- NMPC for structured trajectory tracking under constraints (actuator-aware guidance)
- DRL (PPO) policy for reactive obstacle avoidance and recovery behaviors
- Hybrid supervisor to select the safer control mode in real time
Why it matters (maritime relevance)
- Docking is a high-consequence operation (damage risk, personnel risk, port congestion impact)
- Hybrid control improves robustness + explainability compared with pure learning approaches
- The Unity co-simulation loop supports evidence generation, digital validation, and future HIL-ready workflows
Quick link
- Project overview:
Yacht Autonomous Docking Project/README.md
Folder: Digital Twin Process for Pump System/
A documentation-driven project focused on building and communicating a digital twin process for a pump system—aligned with reliability, monitoring, and lifecycle engineering perspectives common in marine auxiliary systems.
Maritime value
- Supports condition monitoring, fault detection, and maintenance planning
- Maps naturally to shipboard systems: ballast, bilge, cooling water, fuel transfer, firefighting, etc.
- Demonstrates a disciplined approach to model ↔ data ↔ validation workflows
Evidence
- Includes a detailed PDF report in
Digital Twin Process for Pump System/Documents/
Folder: Constraint Programming/Case Study/
Constraint programming models for scheduling a multi-stage production flow (e.g., steel cutting → welding → assembly → finishing). The included Python levels show progressively richer modeling, starting from feasibility.
What’s inside
- Python scripts (
Level1.py…) using constraint programming patterns:- alternative machine selection per stage
- no-overlap constraints per machine
- staged production structure scaffolding
Maritime industry relevance
- Directly applicable to shipyard throughput, block construction sequencing, resource bottlenecks, and lead-time reduction
- Bridges operations research with engineering execution—useful for modern digital shipyard initiatives
Folder: CNC Machining with error detection/
A process and reporting focused project on machining with error detection / lean manufacturing context.
Why it’s included in a maritime portfolio
- Maritime engineering depends on reliable fabrication and inspection workflows
- Demonstrates systems thinking: process monitoring, quality control, and actionable reporting
This repo contains work across multiple languages and engineering environments, including:
- Python (simulation, optimization/constraint models, control/ML pipelines)
- C / C# (systems-level + tooling environments)
- Unity 3D (marine 3D environments)
- Blender (design environment)
- Flexim (process and operations modeling)
- Simcenter Amesim (mechanical engineering modeling)
- Plus supporting documents and artifacts (PDF/PPTX/media)
- Start with the flagship autonomy project:
Yacht Autonomous Docking Project/README.md - Review supporting engineering depth:
- Pump system digital twin documentation
- Shipbuilding scheduling optimization models
- Check project folders for documents, code, and visual evidence (images/videos where included)
Lemuel Hornsby‑Odoi
Marine Engineer — Autonomous Marine Systems • Maneuvering & Control • Digital Twins
MSc Thesis Project work documented in this portfolio (April 2026)