Using non-expert language feedback in learning is crucial to making robots successful in human-centered environments. While language feedback has shown potential to teach robots complex tasks, it also brings challenges: humans leave important context, detail, and clarifications unspoken, making interactive approaches necessary to use the feedback effectively. In this work, we develop an interactive robot learning system that can ask clarifying questions to differentiate between hard and soft task constraints from user verbal feedback. The system uses feedback as either shields (hard constraints) or to shape the reward (soft constraints). We conducted a user study with 24 participants, comparing the use of both hard and soft constraints versus two baseline conditions. We show that participants significantly prefer a system using a combination of both hard and soft constraints, or only using soft constraints, compared to a system using only hard constraints. Qualitative analysis of the participants' interactions with the system revealed common feedback types: spatial, temporal and meta-level. To evaluate the learning performance of the system, we conducted simulated experiments showing that combining both hard and soft constraints performs best in terms of reaching high rewards and finding an efficient solution. Additionally, we provide demonstrations of our system on real robot hardware.

Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can precompute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.

Digital image processing is highlighted in the medical setting for automatic diagnosis of diseases. Glaucoma has no cure, and is the second leading cause of blindness worldwide. Currently there are treatments to prevent vision loss, but the disease will be discovered still in the early stages. Thus, this study aims to develop an automatic detection method of Glaucoma in retinal images. The methodology used in the study was: image acquisition, segmentation of the optic disc area (DO) in retinal images, color feature extraction and entropy in the targeted area and shortly after the selection of attributes. Finally, classification of images was conducted for identifying glaucoma.

Distance learning boosts the development of Web-based software aimed at supporting teaching and learning processes. More recently, information technology became part of pedagogies since their conception, especially for those called “pedagogical architectures” focused mainly on learning and on the processes of creation, discovery and invention. This scenario presented a new challenge: software tools need to facilitate the expression of individual idiosyncrasies and teams as well as incorporate multiple fine-tuned ways of working. It became necessary to design flexible virtual environments that meet the needs of its users, allowing runtime changes in their structures and interfaces. In this paper we present a platform designed under a novel paradigm for designing virtual environments, with flexibility to be combined and to define several virtual environments, runtime changeable without loss of data. As proof of concept, a software platform was developed and used to model and implement a total of 11 well-known virtual environments spanning from usual ones like forum, blog and wiki, to more complex examples. Finally, in order to evaluate the platform's ability to accommodate changes, a case-study was carried out in a real-world event with a class of 10 multipliers (teachers responsible for training other teachers) that reported their use of the platform's resources available so far.