Paintball is a sport in which competing teams try to shoot each
other with guns that fire paint-filled pellets that break on impact,
leaving a colorful mark on the target. It is usually played in an arena
decorated with barriers and other objects that can be used as
cover.
Suppose you are playing paintball in an indoor arena 30 feet wide
and 50 feet long. You are standing near one of the 30 foot walls, and you
suspect that one of your opponents has taken cover nearby. Along the wall,
you see several paint spatters, all the same color, that you think your
opponent fired recently.
The spatters are at 15, 16, 18, and 21 feet, measured from the
lower-left corner of the room. Based on these data, where do you think
your opponent is hiding?
Figure 9-1 shows a diagram of the arena. Using
the lower-left corner of the room as the origin, I denote the unknown
location of the shooter with coordinates α and β, or
alpha and beta. The location of a spatter is labeled
x. The angle the opponent shoots at is
θ or theta.
The Paintball problem is a modified version of the Lighthouse
problem, a common example of Bayesian analysis. My notation follows the
presentation of the problem in D.S. Sivia’s, Data Analysis: a
Bayesian Tutorial, Second Edition (Oxford, 2006).
Figure 9-1. Diagram of the layout for the paintball problem.
The suite
To get ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month, and much more.
