Figure 1: Examples of children's drawings and drawing activity setting.

As Computational Thinking (CT) becomes more prevalent in school curricula, it has reached young children, with growing interest in incorporating CT for children in preschool and kindergarten.As more researchers design systems and curricula to teach CT for this age group, there is a need for assessments to better evaluate and compare approaches to teach CT.In this work-in-progress we present activities with 5-6-year-old children from two classrooms.In these classrooms, children participated in educational robotics (ER) activities using a robot that can be programmed through cards placed on the floor for the robot to read as it moves.We developed a test aimed at evaluating children's ability to program the robot, which we administered after the activities.We discuss lessons learned through the administration of the test, both about the ER activities and the test.

This project focuses on implementing and validating a simple, general, and expandable pipeline for object manipulation. It is implemented as a modular framework that uses 5 configurable modules: camera controller, object detection, object segmentation, grasp proposal, and an executor (arm controller). In the current version, SUMBA implements some of the most widely used methods for each module, and the user can choose any combination of them or implement their own. These solutions are evaluated on a real robot with objects from the YCB Object and Model Set, in a tabletop manipulation task. The best results were obtained by using YOLOv8 both for object detection and segmentation.

It is important to estimate the number of fruits which are going to be harvested, since that information allows farmers to make key decisions regarding the production, distribution and commercialization of crops. In this project, solutions for detecting and counting apples are explored, by using detection and tracking algorithms based on neural networks. Four neural network object detection models were trained: YOLOv5, YOLOv8 and two variants of Faster R-CNN, using public as well as self-made datasets. Three tracking algorithms (StrongSort, ByteTrack and OCSort) were tested using the trained models and labeled video datasets. These datasets were developed with recordings of a real-world orchard and recordings generated with a ROS-based apple field simulator, developed as part of this project. Finally, two fitting models based on linear regression were evaluated in order to minimize prediction errors. As a conclusion of this research, it was found that the YOLOv5 and YOLOv8 performances were better than the other two, achieving a mAP50 of 0.87 and 0.85. In terms of apple tracking, the performance of OCSort stands out, reaching a HOTA metric of 60.24% and running times in the order of 30 ms.

Robotito is an omnidirectional robot designed and developed at Universidad de la República, Uruguay. It is part of a research line in educational robotics aimed at developing free software and open hardware robots for educational use. It was developed in 2018 and has been used in various research projects and educational activities since then. The demo aims to present the robot and its two programming interfaces (tangible and digital interface) to the IDC community to discuss its use in research and education, identify possible extensions and improvements, and encourage international collaborations.

Parkinson's Disease (PD) is a neurological disorder characterized by tremors, difficulty in movement, gait and coordination. Among other symptoms, most patients suffer from Freezing of Gait (FOG): an abrupt halt in gait, usually described by patients as “feet get glued to the ground”. Research has shown that cueing techniques help improve the patient's gait and reduce the number and duration of FOG episodes. Cueing can be defined as using external stimuli which provide temporal or spatial information to facilitate movement initiation and continuation. Although visual stimuli help alleviate FOG, state-of-the-art devices only help with scaling the gait and do not consider other known FOG triggers, such as changes in direction. Our solution (ARFoG) uses a stereo camera and state-of-the-art robotics software to dynamically build a map of the environment, localize the patient and plan trajectories to where they want to go. Once a destination is chosen via voice commands, lines indicating where to step throughout the entire trajectory are shown using Smart Glasses. With this device, the patient's focus is only on the gait movement, avoiding distractions that may trigger FoG episodes.

In recent times, unmanned autonomous sailboats have gained significant relevance due to the energy advantages they offer compared to other vessels, as their main power source is the wind, allowing them to sail for indefinite periods. One of the main challenges these sailboats face is navigating in diverse weather conditions, which has led to the need for developing an adaptive controller capable of handling these situations. This work proposes a solution to this problem by generating an adaptive controller using reinforcement learning combined with imitation learning and utilizing a simulator for training and evaluation. The conducted experiments demonstrate that the generated controllers can navigate in multiple environmental conditions and perform better than the static controller used as a reference. It is worth noting that only reinforcement learning is employed in training the resulting controllers, as imitation learning did not prove beneficial. The code can be found in the repository [1].

Programmable robots designed for preliterate children are one of the options being explored and put into practice for teaching computational thinking skills to children in preschool and kindergarten. Classroom use of these robots may involve use by groups of children due to cost, logistical, and pedagogical reasons. To understand design factors affecting the social use of these robots, we explored the use of three programmable robots with distinctive design characteristics in a kindergarten classroom. Our findings suggest that programmable robot designs that may work well for use by individual children may cause difficulties when shared by groups of children if not all children in the group are able to easily perceive the input (program), output (robot actions), or program state. Based on these design factors we provide recommendations for the design of programmable robots, their evaluation for social use, and for addressing design limitations with support by adult facilitators.

We present a biologically-inspired computational model of the rodent hippocampus based on recent studies of the hippocampus showing that its longitudinal axis is involved in complex spatial navigation. While both poles of the hippocampus, i.e. septal (dorsal) and temporal (ventral), encode spatial information; the septal area has traditionally been attributed more to navigation and action selection; whereas the temporal pole has been more involved with learning and motivation. In this work we hypothesize that the septal-temporal organization of the hippocampus axis also provides a multi-scale spatial representation that may be exploited during complex rodent navigation. To test this hypothesis, we developed a multi-scale model of the hippocampus evaluated it with a simulated rat on a multi-goal task, initially in a simplified environment, and then on a more complex environment where multiple obstacles are introduced. In addition to the hippocampus providing a spatial representation of the environment, the model includes an actor-critic framework for the motivated learning of the different tasks.

Computational thinking is a skill that is considered essential for the future generations. Because of this it should be incorporated into the curricula as soon as possible. An interesting option to work on computational thinking with children is by means of robots. Here, we present Robotito, a robot that can be programmed by arranging its environment, intended to help the development of computational thinking in preschool children. We describe its hardware and software environment, and hierarchical state machines used to implement two modes of interaction with environment-first based on color detection and the second sensible to the surrounding objects. We also present activities that we developed to work on abstraction, generalization, decomposition, algorithmic thinking, and debugging-skills related to computational thinking.

The MateFun project arises with the intention of approaching disciplines that intersect in the field of education, based on the transversality of information and communication technologies with respect to Engineering, Communications, Psychology, Teaching and Pedagogy. In Uruguay, computer courses are increasingly being integrated into curricular content, since learning to program seems to be part of the skills needed for today's people. However, programming in general is absent in mathematics teaching, contrary to the inherent relationship of both. In turn, according to 2015 data from the Educational Monitor of the National Administration of Public Education of Uruguay, Mathematics is the least approved subject of the basic secondary cycle. MateFun is a functional programming language, accessible from a web application, especially aimed to math functions learning. It is intended that through MateFun the learning of programming strengthens the appropriation of the concept of mathematical function, at the same through this project we seek to generate scientific evidence of the transfer or the contributions of programming learning to math. Preliminary data show that the adolescents who experimented with MateFun had similar learning to the control group in mathematical functions, but they also acquired basic knowledge of functional programming.

Grid cells and place cells have shown to play an important role in spatial cognition in rats. While place cells provide global localization from external environment information, grid cells provide a “neural odometry” for path integration from internal vestibular information. In this paper we describe a spatial cognition model integrating grid cells and place cells from behavioral and neurophysiological brain studies in rats. Grid cell firing is generated from a linear oscillatory interference model that includes a reset mechanism to overcome errors in “neural odometry” readings. The model is evaluated in simulation. Future work is discussed including extensions to the model and evaluation under physical robots.