This year’s annual robot roundup is a bit special – there are no LEGO bots! Due to COVID-19, the roundup was forced to be moved online. Instead of the usual Lego bots and black grids, we have Python programs and a virtual board simulator.

​

Unfortunately, our team made a critical mistake in the code due to our misunderstanding of how the simulator worked. This mishap led to the Oracle not receiving the commands given. Instead of following our algorithm, it kept moving right regardless of its surroundings, and eventually into the Wumpus in the very first round.

​

Although our bot didn’t fare so well in practice, we believe that the algorithm we came up with worked just as intended. During our dry runs, it was able successfully to retrieve the gold 100% of the time. I think the Oracle would’ve held its own during the roundup if we were able to effectively implement the algorithm.

​

Our team’s misadventures aside, most of the other bots performed very well, with only two in total falling during the first round. As the board became more and more complicated, the diversity of the bots’ logics became apparent – some took the most efficient route to the gold and back, some took their time venturing most of the board before heading for the gold, and some even went out of their way to kill the Wumpus if possible.