A
bd
el
ba
ki B
o
u
g
u
er
r
a
R
ob
us
t E
xe
cu
tio
n o
f R
ob
ot T
as
k-Pla
ns
:
A K
no
w
led
ge
-b
as
ed A
pp
ro
ac
h
2008 issn 1650-8580 isbn 978-91-7668-610-2Abdelbaki Bouguerra has been a graduate student at the Center for Applied Autonomous Sensor Systems (AASS), Örebro university, Sweden. His research interests include artificial intelligence reasoning and decision making, mobile robotics, and robust execution of robot actions.
Plan execution and monitoring by mobile robots is a complex task as it involves dealing with uncertainty and unexpected situations. Uncertainty is an inherent feature of acting in real-world environments, and it can cause even the best-laid plans to fail. To correctly execute their plans, mobile robots employ execution monitoring procedures that help them detect and deal with unexpected execution situations.
This thesis investigates how advanced forms of artificial intelligence reaso-ning techniques can be used to increase the robustness of plan execution by mobile robots. The thesis proposes to employ semantic domain-knowledge as a source of information to compute and monitor implicit expectations of executing actions correctly.
The thesis also explores how a mobile robot can handle unexpected execu-tion situaexecu-tions using sensor-based artificial intelligence planning.
Örebro Studies in Technology 32 örebro 2008
Doctoral Dissertation
Robust Execution of Robot Task-Plans:
A Knowledge-based Approach
Abdelbaki Bouguerra Technology