Minimalist Mobile Robotics

100

4.

Vision

currently perform will activate only a single module, the details

of this network do not matter. However, one could imagine

cases where the structure of the arbitration network could

produce some useful action, such as selecting the module closest

to the center.

image

of .

workspace

ALIGN

left

APPROACH

ALIGN

right

Figure

4-14.

This picture shows

the

response regions

of the

modules

in

the control system overlaid

on an

actual image

of the

center

of the

arm's

workspace.

The

robot does

not

approach

if

there

is any

danger that

the can

might

move

off the

side

of the

image.

APPROACH and ALIGN form the core of the Seek level of

competence ...