My Capstone Project
Autonomous Sailboat Control System
I'm a passionate engineer specializing in control systems and autonomous robotics. This project showcases the culmination of my work on real-time sensor fusion, optimal path planning, and embedded systems for an autonomous sailboat fleet.
Skills & Tools
Core Expertise
Throughout this project, I’ve leveraged a range of tools and methodologies to design, implement, and test an autonomous sailboat control system.
- Programming: C++, MATLAB, ROS2
- Embedded Systems: Raspberry Pi, Arduino, ESP32
- Control Theories: PID, LQR, Kalman Filters
- Data Handling: Sensor Fusion, Real-Time Processing
- Software Tools: Simulink, Git, Docker
Problem Statement
The Challenge
Global shipping incurs roughly $45B in extra fuel costs annually due to encountering unpredicted ocean waves, contributing nearly 1% of the world’s total GHG emissions. By accurately forecasting wave conditions, routing algorithms can improve efficiency by over 20%, equivalent to eliminating the emissions of 70 million cars. Our solution involves a swarm of autonomous, self-powered ocean sensors that dynamically position themselves to measure wave height, frequency, and direction—inputs vital for newly developed wave-prediction models.
As part of my capstone project, I am specifically focusing on the guidance and navigation systems for each autonomous vessel, coordinating them through swarm mesh networks to optimize data collection and smart routing.
Methodology & Key Challenges
Methodology
To ensure accurate navigation and reliable performance, the project incorporates a comprehensive set of methodologies:
- Utilizing MATLAB/Simulink for initial controller modeling and validation.
- Implementing ROS2 for modular control and real-time data management.
- Incorporating sensor fusion (GPS, IMU, anemometer) to track position and wind data.
Key Challenges
- Real-Time Adaptability: Balancing computational load with the need for rapid adjustments to wind shifts.
- Sensor Reliability: Ensuring robust performance under harsh marine conditions.
- Energy Efficiency: Maintaining low power usage for longer voyages.
- Environmental Factors: Accounting for wave height, currents, and gusts.
Control Algorithm Design
Core Concepts Learned in Control System Development
The autonomous sailing system was developed using a modular control architecture, integrating advanced control techniques to optimize navigation, stability, and efficiency.
Waypoint Management & Optimization
- Dynamic Waypoint Handling:Implemented a queue-based system to manage waypoints efficiently.
- LQR-Based Path Optimization:Applied LQR to dynamically adjust waypoints and avoid inefficient paths.
Path Planning & Sensor Fusion
- Complex Maneuver Planning:Developed algorithms for advanced sailing maneuvers (tacking, jibing, etc.).
- Kalman Filtering:Refined sensor data (GPS, magnetometers, wind sensors), reducing noise and improving accuracy.
Control Theory & Feedback Loops
- PID Control for Rudder Adjustment:Fine-tunes rudder inputs for precise course corrections.
- Drift Estimation with Kalman Filters:Compensates for currents and gusts, improving navigation stability.
Current Progress
Latest Developments
The initial prototypes have successfully completed inland lake tests with promising results for stable heading. A newly integrated sensor suite (9-DOF IMU, high-precision GPS) further improves navigation accuracy. Current work focuses on advanced wind prediction models and dynamic route optimization.
Additional Insights
Collaboration & Personal Growth
Working in a large research lab has taught me the importance of cross-disciplinary communication, agile iteration, and collaborative problem-solving. Throughout this project, I’ve gained experience coordinating with mechanical, electrical, and software engineers to ensure each subsystem meets the overall system requirements.
I’m also actively pursuing publication opportunities to share breakthroughs in wave forecasting and swarm coordination. This process has honed my technical writing and peer collaboration skills, further enhancing my ability to learn quickly and adapt to new methodologies.
Simulation & Field Testing
A significant portion of the development cycle involves running simulations in both calm-water and rough-sea scenarios. This approach allows rapid iteration of the control algorithms before on-water tests. Once the simulated performance meets certain reliability thresholds, real-world trials validate assumptions and calibrate sensor parameters.
Future Goals
Where It’s Heading
- Adaptive AI Integration: Predict wind shifts and optimize sail trim in real-time.
- Oceanic Expedition Trials: Transition from lakes to offshore testing for endurance and resilience.
- Energy Harvesting: Explore solar and wind turbine systems for multi-day journeys.
- Fleet Coordination: Investigate swarm intelligence for multiple autonomous sailboats.
Interested in More Details?
Dive deeper into the project code or feel free to connect with me for collaboration opportunities.