Frustration: a generic mechanism to improve autonomy in robotics.

In this work, we focus on how an emotional controller can be used to modulate robot's behaviors. We present a generic neural architecture, based on an online novelty detection algorithm, that may be able to self-evaluate any sensori-motor strategies. We show how a simple mechanism based on the prediction progress allows the system to monitor its strategies and communicate its disability in deadlock situations. We made several experiments that can account for such properties for two different behaviors (road following and place cells based navigation) in a simulated environment