This paper presents an innovative chatbot architecture designed to support decision-making in the context of mobility data, leveraging recent advancements in Large Language Models (LLMs). As transportation systems and location-based services produce more data, understanding human mobility patterns and providing relevant insights becomes critical for effective decision-making. The chatbot aims to offer a user-friendly tool that allows users to interact with mobility datasets through natural language, preventing the end user from writing complex SQL queries and allowing them to create data visualization on the fly.

Humans learn multisensory eye-hand coordination starting from infancy without supervision. For an example, they learn to track their hands by exploiting various sensory modalities, such as vision and proprioception. This integration occurs as they learn to perceive the world around them and their relationship to it. Most prior work has focused on the role of vision, as it is a primary sensory source for humans. However, it is interesting to study how vision and proprioception interact. We propose a system which combines visual and proprioceptive information to learn the eye-hand coordination skills that enable a robot to fixate its camera gaze on the end effector of its arm. In our model, visual cues are part of the feedback control loop, whereas proprioceptive cues are part of a feedforward control loop. Both controllers, as well as the sensory transform from raw visual information to a neural sensory representation are learned as the robot performs motor babbling movements. Visual information is encoded by sparse coding. The basis functions that emerge are similar to the receptive fields in the human visual cortex. An actor-critic reinforcement learning algorithm is used to drive eye motor neurons fusing visual and proprioceptive cues. We model and test the system in the iCub simulation environment. Our results suggest that these sensory modalities are capable of jointly learning model parameters to perform the tracking task. The evolved policy has characteristics that are qualitatively similar to the human oculomotor plant.

Autonomous robots and humans need to create a coherent 3D representation of their peripersonal space in order to interact with nearby objects. Recent studies in visual neuroscience suggest that the small coordinated head/eye movements that humans continually perform during fixation provides useful depth information. In this work, we mimic such a behavior on a humanoid robot and propose a computational model that extracts depth information without requiring the kinematic model of the robot. First, we show that, during fixational head/eye movements, proprioceptive cues and optic flow lie on a low dimensional subspace that is a function of the depth of the target. Then, we use the generative adaptive subspace self-organizing map (GASSOM) to learn these depth-dependent subspaces. The depth of the target is eventually decoded using a winner-take-all strategy. The proposed model is validated on a simulated model of the iCub robot.

Saccades are fast eye movements that allow humans and robots to bring the visual target in the center of the visual field. Saccades are open loop with respect to the vision system, thus their execution require a precise knowledge of the internal model of the oculomotor system. In this work, we modeled the saccade control, taking inspiration from the recurrent loops between the cerebellum and the brainstem. In this model, the brainstem acts as a fixed-inverse model of the oculomotor system, while the cerebellum acts as an adaptive element that learns the internal model of the oculomotor system. The adaptive filter is implemented using a state-of-the-art neural network, called I-SSGPR. The proposed approach, namely recurrent architecture, was validated through experiments performed both in simulation and on an antropomorphic robotic head. Moreover, we compared the recurrent architecture with another model of the cerebellum, the feedback error learning. Achieved results show that the recurrent architecture outperforms the feedback error learning in terms of accuracy and insensitivity to the choice of the feedback controller.

Autonomous robots need to make sense of their surroundings, recognize objects, detect and, possibly, identify the people around them. Visual perception would be ideally suited for these tasks. Yet, visual perception remains one of the limiting factors of modern robotics: artificial vision systems tend to perform poorly in normal environments, where the scene and the illumination conditions are unpredictable. Evolution has faced similar problems, leading to surprisingly accurate visual capabilities, even in species with very small brains. We argue that the success of biological perception systems relies on three fundamental computational principles: (a) the continual coupling of perception and behavior; (b) the resulting emergence of multimodal cues; (c) and their efficient integration. Building on these computational principles, we describe a humanoid robot that emulates the dynamic strategy by which humans examine a visual scene. The proprioceptive and visual depth cues resulting from this strategy are integrated in statistically optimal manner into a unified representation. We show that this approach yields accurate and robust 3D representations of the observed scene.

Active eye movements can be exploited to learn the oculomotor command that allows a humanoid head to gaze at a visual target. In the field of adaptive robotics this transformation is commonly encoded by means of neural networks that are trained using ballistic eye movements. In this work, we encode the visuo-oculomotor transformation using radial basis function networks and we derive the equations to use such a transformation in visual servo control tasks. We study how the choice of the input (monocular vs binocular) influences the execution of ballistic and visually-guided eye movements. Analytic analysis and experimental results on a simulated model of the robot show that the proposed framework can be employed to execute both kind of movements.

Image processing has frequently to deal with thin structures that hold the information required. In natural images, there is a wide variability of conditions and the borders of objects are not always well defined. Progressive scan, an algorithm derived from fuzzy-connectedness theory, is described here as a specific case of more general /spl chi/-connectedness. Exploiting some analogies to the mechanical world, it allows one to track thin structures and consequently, is suitable for river, motorway or vessel extraction. In the computation of fuzzy connectedness, a widely used approach is based on an adaptive growing mechanism that follows the best paths starting from a reference seed point. The new algorithm uses three or more seed points in order to better drive that approach along the structure of interest. Results of the algorithm are the best path along an object and a connectivity map. The algorithm can deal with heterogeneous images ranging from medical to remote sensed ones.