Remote Agent: to boldly go where no AI system has gone before
Artificial Intelligence - Special issue: artificial intelligence 40 years later
Introduction to the Theory of Computation
Introduction to the Theory of Computation
Lessons learned from autonomous sciencecraft experiment
Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems
IEEE Intelligent Systems
Mixed-initiative activity planning for Mars rovers
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Hi-index | 0.00 |
In addition to its utility in terrestrial-based applications, Automated Planning and Scheduling (P&S) has had a growing impact on space exploration. Such applications require an influx of new technologies to improve performance while not comprimising safety. As a result, a reliable method to rapidly assess the effectiveness of new P&S algorithms would be desirable to ensure the fulfillment of of all software requirements. This paper introduces RoBen, a mission-independent benchmarking tool that provides a standard framework for the evaluation and comparison of P&S algorithms. RoBen considers metrics derived from the model (the system on which the P&S algorithm will operate) as well as user input (e.g., desired problem complexity) to automatically generate relevant problems for quality assessment. A thorough description of the algorithms and metrics used in RoBen is provided, along with the preliminary test results of a P&S algorithm solving RoBen-generated problems.