Wind Tunnel Valve Mobot
An autonomous robot designed to navigate a tunnel course and activate a valve using sensors and mechanical control
00
Problem
The challenge was to design and build a fully autonomous mobile robot capable of navigating a tunnel-like course, counting ground lines, following walls to reach a valve, and hitting the valve precisely at the end. The system needed to remain accurate despite variations in starting orientation, surface friction, and sensor noise, while maintaining reliable movement and stability .
Solution
I created a compact, two-wheeled robot equipped with line sensors, an ultrasonic sensor, a gyroscope, and dual servo-driven motors. Our control logic allowed it to count six lines at the start, adjust course mid-tunnel using wall-following with the ultrasonic sensor, and realign orientation using gyroscopic feedback before striking the valve with a mounted servo arm. This approach allowed the robot to complete the course reliably despite environmental inconsistencies .
We began by dividing the robot into three engineering subsystems: control systems, solid mechanics, and thermal-fluids-energy. Early trials revealed significant variation in robot pathing after sharp turns, which often misaligned it with the tunnel. We solved this by coding a wall-following algorithm using ultrasonic feedback, letting the robot self-correct regardless of starting angle.
We then optimized the robot’s body to balance stability and speed. Structural analysis helped us distribute weight over the drive wheels to maintain traction, while keeping the frame light for efficient movement. On the thermal-fluids side, we modeled various front-end geometries (cowcatcher, parachute, pyramid) in Ansys Discovery to minimize drag and improve energy efficiency.
In final integration tests, the robot consistently counted lines, navigated to the valve, reversed to align, and struck the valve with high accuracy. This outcome showed how multiple engineering principles could work together through careful system integration and iterative refinement .
01
Fluid simulations were run in Ansys Discovery to evaluate different front-end shield geometries and minimize aerodynamic drag. By visualizing airflow patterns and measuring pressure drop and velocity, we identified the cowcatcher shape as the most efficient design, achieving lower resistance and smoother airflow compared to other tested profiles.
02
The front-end shield of the robot was fabricated using FDM 3D printing on an Ender machine. This lightweight component was designed to minimize aerodynamic drag while protecting onboard electronics, and additive manufacturing allowed rapid iteration to refine its geometry and fit.
03
The robot successfully navigated the course using its line-following and wall-tracking system, maintaining accuracy even with visual distractions on the path. The front-end shield helped it stay stable against wind disturbance during wind tunnel tests, and it ultimately reached the end of the course and activated the valve as intended.
+
