Adriano Manfré

Creation and cognition for humanoid live dancing

Abstract Computational creativity in dancing is a recent and challenging research field in Artificial Intelligence and Robotics. We present a cognitive architecture embodied in a humanoid robot capable to create and perform dances driven by the perception of music. The humanoid robot is able to suitably move, to react to human mate dancers and to generate novel and appropriate sequences of movements. The approach is based on a cognitive architecture that integrates Hidden Markov Models and Genetic Algorithms. The system has been implemented on a NAO robot and tested in public setting-up live performances, obtaining positive feedbacks from the audience.

Learning by Demonstration for a Dancing Robot within a Computational Creativity Framework

The paper presents a system that learns a set of movements for a creative dancing robot. A human user only dances in front of an 3D camera, and automatically the acquisition system segments the acquired sequence of postures depending on the detected music beat and rhythm. A clustering phase allows the system to group the identified actions in 20 classes, defining the set of movements that is typical of a given person. Analysis of the k-mean algorithm outcomes using different distances is reported. The human postures are translated in the corresponding robot joints configurations and are used to compose dance choreographies creatively. A cognitive architecture developed in previous works drives the process of dance creation. Experimentation shows the sets of movements derived from human users with different dance skills. Audience evaluates the robot performances based on these sets, and results are coherent with the quality and richness of the acquired movements.

A Personal Intelligent Coach for Smart Embodied Learning Environments

Within a Smart Learning Environment (SLE) learners are involved in a new learning process tailored to create a continuum of education by extending the current educational formal settings to real-life informal learning context. The goal of this paper is to describe the Cognitive Architecture (CA) of a Personal Intelligent Coach able to manage learning tasks and interactions within a complex Smart Learning Environment (SLE). PICo has two possible embodiments: humanoid robot, and an avatar on mobile device. We argue that the proposed intelligent coach can adapt to the contents, to the students needs and can evolve its strategies according the learning process.

Move Your Mind: Creative Dancing Humanoids as Support to STEAM Activities

Educational activities based on dance can support interest in comprehension of concepts from maths, geometry, physics, bio-mechanics and computational thinking. In this work, we discuss a possible use of a dancing humanoid robot as an innovative technology to support and enhance STEAM learning activities.

Improving Spatial Reasoning by Interacting with a Humanoid Robot

This paper analyzes the connection between spatial reasoning and STEM education from the point of view of embodied theories of cognition. A new learning model based on the use of a humanoid robot is presented with the aim of teaching and learning basic STEM concepts in a fruitful and engaging fashion.

Robot Navigation Based on an Artificial Somatosensorial System

An artificial somatosensory system processes robot’s perceptions by mean of suitable soft sensors. The robot moves in a real and complex environment, and the physical sensing of it causes a positive or negative reaction. A global wellness function drives the robot’s movements and constitutes a basis to compute the motivation of a cognitive architecture. The paper presents preliminary experimentations and explains the influence of the parameters on the robot behavior and personality. Pepper freely moves in an office environment searching for people to engage. The robot searches for a safe path, avoiding obstacles and aiming to explore a significant part of a known space by an approximative map stored in its long term memory (LTM). The short-term memory (STM) stores somatosensory values related to perceptions considered relevant for the navigation task. The collection of previous navigation experiences allows the robot to memorize on the map places that have positive (or negative) effects on robot’s wellness state. The robot could reach the places labeled as negative, but it needs some positive counter effects to contrast its reluctance.