MEASURING THE EFFECTIVENESS OF REINFORCEMENT LEARNING FOR BEHAVIOR-BASED ROBOTS

We explore She use of behavior-based architectures within the context of reinforcement learning and examine the effects of using different b ehavior-based architectures on the ability to learn correctly and effi ciently the task at hand. in particular, we study the task of learning to push boxes in a simulated two-dimensional environment originally p roposed by Mahadevan and Connell (1992). We examine issues such as eff ectiveness of learning, flexibility of the learning method to adapt to new environments, and effect of the behavior architecture on the abil ity to learn, and we report results obtained on a large number of simu lation runs.