Page 319 - Introduction to AI Robotics

302

b. 8 Multi-agents

c.
a.

Figure 8.3 RoboCup soccer competition: a.) Overhead view of the RoboCup playing
field and teams for the small-sized league (overhead camera is not shown), b.) Mid-
sized league, and c.) Legged league using Sony Aibo robots. (Photographs c
The
RoboCup Federation 1999. All rights reserved.)

sensing for reflexive obstacle avoidance. In the RoboCup case, the robots
must have a set of basic tactical behaviors (see Ch. 7), but may either receive
strategic commands from the central computer or have on-board strategic
behaviors. This type of control is conceptually equivalent to the Hybrid
Reactive-Deliberative Paradigm, where the reactive layer physically resides
on the robot and the deliberative layer resides on the central workstation.
Distributed control is more natural for soccer playing than centralized
control, because each player reacts independently to the situation. An ex-
ample of distributed control in robot soccer playing is the mid-sized league
in RoboCup. Notice that in robot soccer the robots are inherently hetero-
geneous. Although they may be physically the same, each robot is pro-
grammed with a different role, most especially Goalie, Striker, and Defender.