In this paper, we consider the important problem of safe exploration in reinforcement learning. While reinforcement learning is well-suited to domains with complex transition dynamics and high-dimensional state-action spaces, an additional challenge is posed by the need for safe and efficient exploration. Traditional exploration techniques are not particularly useful for solving dangerous tasks, where the trial and error process may lead to the selection of actions whose execution in some states may result in damage to the learning system (or any other system). Consequently, when an agent begins an interaction with a dangerous and high-dimensional state-action space, an important question arises; namely, that of how to avoid (or at least minimize) damage caused by the exploration of the state-action space. We introduce the PI-SRL algorithm which safely improves suboptimal albeit robust behaviors for continuous state and action control tasks and which efficiently learns from the experience gained from the environment. We evaluate the proposed method in four complex tasks: automatic car parking, pole-balancing, helicopter hovering, and business management.

Recent discoveries in automated planning are broadening the scope of planners, from toy problems to real applications. However, applying automated planners to real-world problems is far from simple. On the one hand, the definition of accurate action models for planning is still a bottleneck. On the other hand, off-the-shelf planners fail to scale-up and to provide good solutions in many domains. In these problematic domains, planners can exploit domain-specific control knowledge to improve their performance in terms of both speed and quality of the solutions. However, manual definition of control knowledge is quite difficult. This paper reviews recent techniques in machine learning for the automatic definition of planning knowledge. It has been organized according to the target of the learning process: automatic definition of planning action models and automatic definition of planning control knowledge. In addition, the paper reviews the advances in the related field of reinforcement learning.

NAOTherapist is a cognitive robotic architecture whose main goal is to develop non-contact upper-limb rehabilitation sessions autonomously with a social robot for patients with physical impairments. In order to achieve a fluent interaction and an active engagement with the patients, the system should be able to adapt by itself in accordance with the perceived environment. In this paper, we describe the interaction mechanisms that are necessary to supervise and help the patient to carry out the prescribed exercises correctly. We also provide an evaluation focused on the child-robot interaction of the robotic platform with a large number of schoolchildren and the experience of a first contact with three pediatric rehabilitation patients. The results presented are obtained through questionnaires, video analysis and system logs, and have proven to be consistent with the hypotheses proposed in this work.