Bomb Retrieval Robot.
ROBOTICS / DATA
UC BERKELEY, 2013-2014
TEAM: ZACH MULDER, CHRIS AU
Designing a model and user interface
for bomb retrieval in an unknown maze
Imagine that you have 45 minutes to retrieve two bombs in unknown space with a robot.
The goal of this project was to design a robot (using the Mindstorms NXT kit) and interface to allow for effective bomb retrieval in an unknown space. The robot was placed in a constructed maze with two “bombs”. Two light beacons were available at both sides of the maze. Using the light beacons, the robot needed to triangulate its position and send it back to a designed interface. Based on this location feedback, the team maneuvered through the maze to find the bombs while sitting in a different room.
A shortest path algorithm was implemented to calculate an optimal path to a destination in a grid that has some blocked intersections, and these calculations were updated as new blocked intersections were discovered using the ultrasonic sensor. Experimental work was conducted to calibrate sensors and determine location with greater accuracy.
- analysis of sensor sensitivities through experimentation
- design and implementation of of algorithmic approach with team
- user-interface design for quick and effective decision-making during test run
Java | sensors | Lego NXT Mindstorms | user-interface design | modeling | data analysis