Recent advances in combining deep learning and Reinforcement Learning have shown a promising path for designing new control agents that can learn optimal policies for challenging control tasks. These new methods address the main limitations of conventional Reinforcement Learning methods such as customized feature engineering and small action/state space dimension requirements. In this paper, we...

What do I need to say to convince you to do something? This is an important question for an autonomous agent deciding whom to approach for a resource or for an action to be done. Were similar requests granted from similar agents in similar circumstances? What arguments were most persuasive? What are the costs involved in putting certain arguments forward? In this paper we present an agent decis...

Research regarding autonomous learning shows that freeplay does not result in optimal learning. Combining scenario-based training with intelligent agent technology offers the possibility to create autonomous training enriched with automated adaptive support delivered by a director agent. We conducted an experiment to investigate whether directing training scenarios improves the quality of train...

In Chapter 17 we started the discussion about autonomous agents as part of the situated cognition approach in AI, and we develop this further here focussing on learning. An agent is autonomous when it is able to function on its own in a complex and changing environment. We will call an agent adaptive when, in addition, it is able to improve its behaviour to make it (more) appropriate for its en...

Autonomous navigation by a mobile robot through natural, unstructured terrain is one of the premier challenges in field robotics. The DARPA UPI program was tasked with advancing the state of the art in robust autonomous performance through challenging and widely varying environments. In order to accomplish this goal, machine learning techniques were heavily utilized to provide robust and adapti...

This thesis is devoted to the study of algorithms for early perceptual learning for an autonomous agent in the presence of feedback. In the framework of associative perceptual learning with indirect supervision, three learning techniques are examined in detail: • short-term on-line memory-based model learning; • long-term on-line distribution-based statistical estimation; • mixed onand off-line...

Developmental robotics is concerned with the design of algorithms that promote robot adaptation and learning through qualitative growth of behaviour and increasing levels of competence. This paper uses ideas and inspiration from early infant psychology (up to 3 months of age) to examine how robot systems could discover the structure of their local sensory-motor spaces and learn how to coordinat...

In order to achieve an autonomous system which can adaptively behave through learning in the real world, we constructed a distributed autonomous swimming robot that consisted of mechanically linked multi-agent and adopted adaptive oscillator method that was developed as a general decision making for distributed autonomous systems (DASs). One of the our aims by using this system is to verify whe...

mobile robot navigation is one of the basic problems in robotics. in this paper, a new approachis proposed for autonomous mobile robot navigation in an unknown environment. the proposedapproach is based on learning virtual parallel paths that propel the mobile robot toward the trackusing a multi-layer, feed-forward neural network. for training, a human operator navigates themobile robot in some...