TrainWreck

side view
top view

Inspiration

Winter Olympic events like the biathlon highlight a rare balance of speed, precision, and composure under pressure. We were inspired by how athletes seamlessly switch between endurance-based movement and high-accuracy targeting, often in harsh and unpredictable conditions. TrainWreck takes this idea into robotics—challenging a robot to navigate complex terrain, make strategic decisions, and execute precise actions, all within a competitive environment.

What it does

TrainWreck is an autonomous robot designed to complete a multi-section challenge inspired by Winter Olympic events. It navigates a course, collects boxes to unlock optional paths, and strategically chooses which sections to attempt. In the Target Shooting section, the robot climbs ramps, uses color sensing to locate the center of a target, and launches a ball toward a scoring zone. In the Obstacle Course section, the robot follows a winding path filled with sharp turns and obstacles, optimizing for speed and smooth movement. The robot returns to its starting position to complete the run.

How we built it

We built TrainWreck by integrating mechanical design, electronics, and software control systems:

Mechanical design for stability on ramps, reliable box pickup, and consistent ball launching
Sensors for color detection, line following, and obstacle awareness
Control algorithms to handle navigation, decision-making, and precise alignment at the target
Modular code structure that allowed us to re-upload and refine logic between sections
Iterative testing to tune movement speed, turning accuracy, and launch consistency Each subsystem was designed to work together under time and performance constraints.

Challenges we ran into

Achieving consistent color-based navigation under varying lighting conditions
Balancing speed vs. accuracy, especially when approaching the target center
Ensuring the ball launcher delivered repeatable results without bouncing off walls
Managing limited time to test edge cases like ramp transitions and sharp turns
Coordinating hardware and software debugging simultaneously Many issues only appeared under full-course conditions, requiring rapid iteration.

Accomplishments that we're proud of

Successfully navigating both optional sections within a single run
Accurately locating the target center using color cues
Launching the ball with controlled direction and distance
Designing a robot that remained stable on both straight and curved ramps
Building a system that rewarded strategy, not just raw speed Most importantly, TrainWreck performed reliably under competition pressure.

What we learned

Small mechanical adjustments can dramatically affect software performance
Sensor noise must be expected—and planned for—in real-world environments
Modular code makes rapid iteration and debugging much easier
Strategic decision-making is just as critical as technical execution
Robotics challenges reward resilience, adaptability, and teamwork This project reinforced how tightly engineering disciplines are interconnected.

What's next for TrainWreck

Improve target-shooting accuracy with closed-loop feedback
Optimize path-planning to dynamically choose the highest-scoring route
Enhance obstacle detection to reduce collision risk at higher speeds
Refine the launcher for more predictable ball trajectories
Add telemetry and logging for deeper performance analysis TrainWreck is just getting started. We see it as a foundation for even smarter, more competitive autonomous systems.