Page 358 - Sensing, Intelligence, Motion : How Robots and Humans Move in an Unstructured World

PRELIMINARY OBSERVATIONS 333

the walker can only see the surrounding walls. The visitor may start inside or
outside of the labyrinth and attempt to reach the center, or locate a “treasure”
inside the labyrinth, or find an exit from it.

Moving with Complete Information. If one has a bird’s-eye view of the
whole labyrinth, this makes the task much easier. To study human performance
in motion planning consistently, we start with this simpler task. Consider the
bird’s-eye view of a labyrinth shown in Figure 7.1. Imagine you are handed
this picture and asked to produce in it a collision-free path from the position S
(Start) to the target point T . One way to accomplish this test is with the help of
computer. You sit in front of the computer screen, which shows the labyrinth,
Figure 7.1. Starting at point S, you move the cursor on the screen using the
computer mouse, trying to get to T while not banging into the labyrinth walls.
At all times you see the labyrinth, points S and T , and your own position in the
labyrinth as shown by the cursor. For future analysis, your whole path is stored
in the computer’s memory.
If you are a typical labyrinth explorer, you will likely study the labyrinth
for 10–15 seconds, and think of a more or less complete path even before you
start walking. Then you quickly execute the path on the screen. Your path will

T

S

Figure 7.1 A two-dimensional labyrinth. The goal is to proceed from point S to point T .