guessing where something is, then betting on thousands of ghosts until one guesses right.
means a statistical technique that estimates a hidden, changing state by simulating many random possibilities and weighting them by how well they match reality.
from grew out of 1950s sequential monte carlo ideas, but the modern algorithm arrived in 1993 when researchers used it to track radar signals bouncing around noisy targets.
robot localization — monte carlo localization lets robots find themselves on a map
apollo-era tracking — precursor ideas tracked spacecraft trajectories through noisy telemetry
financial volatility models — traders estimate hidden market states from noisy price data
self-driving perception — autonomous cars fuse lidar and camera noise into one belief