This is fantastic — in 1950 Claude Shannon, the father of information theory, designed a mechanical mouse that could explore an arbitrary maze, reliably find a path to the goal, remember it, and then recognize and adapt to changes. He called it Theseus. The mouse itself is only a copper-whiskered bar magnet on three wheels; it’s motivated by an electromagnet under the maze floor, driven by a pair of motors. The thinking is done by telephone relays following a topology theorem.
In 1977, when the editors of IEEE Spectrum challenged their readers to design a self-contained micromouse that could solve a maze through trial and error, Shannon took Theseus down from his attic and put him on display beside his descendants, which could now accomplish the same task using their own onboard processors. Computers were “not up to the human level yet,” he said, but “it is certainly plausible to me that in a few decades machines will be beyond humans.”