by F Stulp and M Beetz
Abstract:
One of the main challenges in motor control is expressing high-level goals in terms of low-level actions. To do so effectively, motor control systems must reason about actions at different levels of abstraction. Grounding high-level plans in low-level actions is essential semantic knowledge for plan-based control of real robots. We present a robot control system that uses declarative, procedural and predictive to generate, execute and optimize plans. Declarative knowledge is represented in PDDL, durative actions constitute procedural knowledge, and predictive knowledge is learned by observing action executions. We demonstrate how learned predictive knowledge enables robots to autonomously optimize plan execution with respect to execution duration and robustness in real-time. The approach is evaluated in two different robotic domains.
Reference:
Combining Declarative, Procedural and Predictive Knowledge to Generate and Execute Robot Plans Efficiently and Robustly (F Stulp and M Beetz), In Robotics and Autonomous Systems Journal (Special Issue on Semantic Knowledge), 2008.
Bibtex Entry:
@article{stulp_combining_2008,
author = {F Stulp and M Beetz},
title = {Combining Declarative, Procedural and Predictive Knowledge to Generate
and Execute Robot Plans Efficiently and Robustly},
journal = {Robotics and Autonomous Systems Journal (Special Issue on Semantic
Knowledge)},
year = {2008},
abstract = {One of the main challenges in motor control is expressing high-level
goals in terms of low-level actions. To do so effectively, motor
control systems must reason about actions at different levels of
abstraction. Grounding high-level plans in low-level actions is essential
semantic knowledge for plan-based control of real robots. We present
a robot control system that uses declarative, procedural and predictive
to generate, execute and optimize plans. Declarative knowledge is
represented in {PDDL}, durative actions constitute procedural knowledge,
and predictive knowledge is learned by observing action executions.
We demonstrate how learned predictive knowledge enables robots to
autonomously optimize plan execution with respect to execution duration
and robustness in real-time. The approach is evaluated in two different
robotic domains.},
}