Toward a Test Environment for Autonomous Controllers

In Proceedings of the 5th Italian Workshop on Planning and Scheduling (IPS 2013), pp. 3-10. Turin, Italy, December 4th, 2013

In the last two decades, an increasing attention has been dedicated on the use of high level task planning in robotic control, aiming to deploy advanced robotics systems in challenging scenarios where a high autonomy degree is required. Nevertheless, an interesting open problem in the literature is the lack of a well defined methodology for approaching the design of deliberative systems and for fairly comparing different approaches to deliberation. This paper presents the general idea of an environment for facilitating knowledge engineering for autonomy and in particular to facilitate accurate experiments on planning and execution systems for robotics. It discusses features of the On-Ground Autonomy Test Environment (OGATE), a general testbench for interfacing deliberative modules. In particular we present features of an initial instance of such system built to support the GOAC robotic software.