Page 297 - Introduction to AI Robotics

280
7
The Hybrid Deliberative/Reactive Paradigm
devoted to maintaining an accurate model of the world based on the robot’s
sensors and assigning symbolic labels to regions. Saphira also divides the
deliberation activities among software agents. This provides a high degree
of flexibility. Since software agents are independent, they don’t even have to
run on-board the robot they’re controlling. In the 1996 AAAI Mobile Robot
Competition, 78 robots running Saphira actually had some of their planning
reside on a local workstation, transmitted through a radio link. 62 This will be
covered inmore detail inCh. 8.
The reactive component of Saphira consists of behaviors. The behaviors
VIRTUAL SENSOR extract virtual sensor inputs from the central world model, the Local Percep-
tual Space. The behavioral output is fuzzy rules, which are fused using fuzzy
logic into a velocity and steer command. Fuzzy logic turns out to be a very
natural way of fusing competing demands, and is less ad hoc than Boolean
logic rules (e.g., “if x and y but not z, then turn left”). The behaviors are man-
aged by the plan execution of the planned navigation tasks. The fuzzy logic
mechanism for combining behaviors produces essentially the same results
as a potential field methodology, as described by Konolige and Myers. 77 The
Local Perceptual Space can improve the quality of the robot’s overall behav-
ior because it can smooth out sensor errors. Although this central processing
introduces a computational penalty, the increases in processor power and
clock speeds have made the computational costs acceptable.
The table below summarizes Saphira in terms of the common components
and style of emergent behavior:

Saphira
Sequencer Agent Topological planner, Navigation Tasks
Resource Manager PRS-lite
Cartographer LPS
Mission Planner PRS-lite
Performance Monitoring Agent PRS-lite
Emergent behavior Behaviors fused with fuzzy logic

7.6.2 Task Control Architecture (TCA)

TASK CONTROL Reid Simmon’s Task Control Architecture (TCA) has been used extensively by
ARCHITECTURE (TCA) robots designed for NASA, including Ambler, Dante (Fig. 7.11), and ser-
vice robots. It is also the intelligence inside Xavier (shown in Fig. 7.11, the
Carnegie Mellon University robot which is accessible over the web). Xavier
has has traveled autonomously over 210 kilometers in the hallways at CMU