English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Over the past decades, we havewitnessed amassive utilization of robots in industrial settings. The use of robotic manipulators with high precision and repeatability were fundamental for the increase of efficiency of many manufacturing processes and mass production. Due to the complete predictability of the factory environment, each robot is programmed a priori, and repeats the same set of movements for weeks or even months. In spite of the limited sensory information available, these robots can complete their tasks in an extremely efficient manner. Recent progress in computer vision, robot navigation, estimation theory, reasoning and computing have afforded engineers with the tools to build more flexible robots able to “live” in more unpredictable environments and interact with (non-technical) humans in a rich and sophisticated manner. In the long-run, this research effort will contribute to the ubiquitous presence of robots in our homes and offices to assist, entertain and work together with humans. For this to become true, there are two very challenging questions that need to be addressed. Firstly, there is the issue of complexity, as such systems are likely to be much more complex than industrial robots with regard to the number of motor and perceptual degrees of freedom and will have to operate in a highly unpredictable environment. Secondly, there is the question of programming or commanding these robots, as they will interact over extended periods of time with non-technical users, who are not expected to write sophisticated computer programs each time the robot is required to perform a new task. As some of these challenges are common to many biological social species, we will borrow inspiration from biology, developmental psychology and neuroscience in our approach. Many animals can be seen as complex and flexible systems, able to learn and adapt through a simultaneousmotor, sensorial and cognitive development. Studies in developmental psychology clearly show how human infants progressively learn how to acquire more sophisticated skills over time, through interaction with their own body, the environment and caretakers. At birth, many motor and perceptual capabilities are not yet in place and the acquisition of some skills is a pre-requisite to acquire more complex ones. It would otherwise be very difficult to handle the overwhelming flow of sensory-motor information. In this work, we adopt the approach of developmental robotics Fitzpatrick et al. (2003); Lungarella & Pfeifer (2001); Metta, Sandini, Natale, Manzotti & Panerai (2001) to build systems 28

Topological Mapping and Navigation Using a Developmental Learning Approach Based on Imitation through Sensory-Motor Maps