Kalman Filter in GPS tracking

Imagine you are trying to track the position of a moving car using noisy GPS data, how would you explain the Kalman filter to someone with no technical background?

Use a real-world analogy to make your explanation accessible.