Frustration as a way toward autonomy and self-improvement in robotic navigation

Autonomy and self-improvement capabilities are still challenging in the field of robotics. Allowing a robot to autonomously navigate in wide and unknown environments not only requires a set of robust strategies to cope with miscellaneous situations, but also needs mechanisms of self-assessment for guiding learning and for monitoring strategies. Monitoring strategies requires feedbacks on the behavior's quality, from a given fitness system in order to take correct decisions. In this work, we focus on how an emotional controller can be used to modulate robot behaviors. Following an incremental and constructivist approach, we present a generic neural architecture, based on an online novelty detection algorithm that may be able to evaluate any sensory-motor strategies. This architecture learns contingencies between sensations and actions, giving the expected sensation from the past perception. Prediction error, coming from surprising events, provides a direct measure of the quality of the underlying sensory-motor contingencies involved. We show how a simple emotional controller based on the prediction progress allows the system to regulate its behavior to solve complex navigation tasks and to communicate its disability in deadlock situations. We propose that this model could be a key structure toward self-monitoring. We made several experiments that can account for such properties with different behaviors (road following and place cells based navigation).