A5039614564- Video Surveillance and Tracking Methods
- Robotic Path Planning Algorithms
- Autonomous Vehicle Technology and Safety
- Logic, Reasoning, and Knowledge
- AI-based Problem Solving and Planning
- Human Pose and Action Recognition
- Robotics and Sensor-Based Localization
- Anomaly Detection Techniques and Applications
- Evacuation and Crowd Dynamics
- Advanced Vision and Imaging
- Indoor and Outdoor Localization Technologies
- Data Mining Algorithms and Applications
- IoT-based Smart Home Systems
- Computability, Logic, AI Algorithms
- Rough Sets and Fuzzy Logic
- Radio Wave Propagation Studies
- Context-Aware Activity Recognition Systems
- Millimeter-Wave Propagation and Modeling
- Complex Systems and Decision Making
- Robot Manipulation and Learning
- Bayesian Modeling and Causal Inference
- Distributed Control Multi-Agent Systems
University of Trento
2018-2023
Dalle Molle Institute for Artificial Intelligence Research
2017
Generating accurate and efficient predictions for the motion of humans present in scene is key to development effective planning algorithms robots moving promiscuous areas, where wrong decisions could generate safety hazard or simply make presence robot "socially" unacceptable. Our approach predict human based on a neural network peculiar kind. Contrary conventional deep networks, our embeds its structure popular Social Force Model, dynamic equation describing physical terms. This choice...
The accurate localization of robots in indoor environments is a well-known measurement problem, which becomes particularly critical crowded scenarios, where collision with obstacles, objects, and human operators very high. In this article, the problem solved through an optimal path planning strategy that keeps into account ranging uncertainty from given set wireless anchors (e.g., ultra wideband (UWB) transceivers) are used to track position robot known reference frame multilateration....
Among the manifold wireless technologies recently adopted for indoor localization and tracking, radars based on phased-array transceivers at 60 GHz are gaining momentum. The main advantages of this technology are: high accuracy, good ability to track multiple target with a low computation burden preservation privacy. Despite growing commercial success low-cost radar platforms, accurate studies evaluate their tracking performance not frequent in literature, reasons being policies that prevent...
Service robots are becoming more and pervasive in natural environments where human beings actually carry out daily duties. One of the main enabler these scenario is capability such a system to understand positions motions people moving its surroundings. In this paper, we will investigate effectiveness tracking that uses stereo camera as sensor then use gathered data an estimation algorithm continuously track their position environment. Experiments laboratory carried characterise evaluate...
Service robots are increasingly applied in real life environments populated with human beings. In such a challenging scenario, the autonomous have to avoid collision "natural" way, that is execute trajectories would follow. This goal can be efficiently tackled if sufficiently descriptive motion model available, order predict future pedestrian behaviour and hence safely planning correct route. this paper, we move first step towards able describe certain extent nonverbal negotiation of spaces...
Abstract One of the most important barriers towards a widespread use mobile robots in unstructured, human populated and possibly a-priori unknown work environments is ability to plan safe path. In this paper, we propose delegate activity operator that walks front robot marking with her/his footsteps path be followed. The implementation approach requires high degree robustness locating specific person followed (the path-finder ). We three phases fulfil goal: 1. Identification tracking image...
One of the most important barriers toward a widespread use mobile robots in unstructured and human populated work environments is ability to plan safe path. In this paper, we propose delegate activity operator that walks front robot marking with her/his footsteps path be followed. The implementation approach requires high degree robustness locating specific person followed (the leader). We three phase fulfil goal: 1. identification tracking image space, 2. sensor fusion between camera data...
Abstract Robot-assisted navigation is a perfect example of class applications requiring flexible control approaches. When the human reliable, robot should concede space to their initiative. makes inappropriate choices controller kick-in guiding them towards safer paths. Shared authority way achieve this behaviour by deciding online how much be given and retained robot. An open problem evaluate appropriateness human’s choices. One possible consider deviation from an ideal path computed This...