Adriana Draghici

Multi-processor architecture for a versatile autonomous robot

This paper presents a versatile, multi-role robotic platform optimized for advanced motion in line-following scenarios with application to robotic contests, education and specialized tasks. Generally, academic robotic platforms either provide only basic motion features, or are too specialized. Our platforms goal is to be sufficiently versatile to be adapted easily to any task, but to provide performance in common scenarios. For this aim, our modular system is comprised of a Mecanum-wheel based omnidirectional motion system with a dense array of line sensors. The system comprises other sensors and actuators, which can be replaced or augmented with more specialized sensors and mechanical systems for each application. We demonstrate our platform in a contest scenario.

A Survey of Techniques for Automatically Sensing the Behavior of a Crowd

Crowd-centric research is receiving increasingly more attention as datasets on crowd behavior are becoming readily available. We have come to a point where many of the models on pedestrian analytics introduced in the last decade, which have mostly not been validated, can now be tested using real-world datasets. In this survey, we concentrate exclusively on automatically gathering such datasets, which we refer to as sensing the behavior of pedestrians. We roughly distinguish two approaches: one that requires users to explicitly use local applications and wearables, and one that scans the presence of handheld devices such as smartphones. We come to the conclusion that despite the numerous reports in popular media, relatively few groups have been looking into practical solutions for sensing pedestrian behavior. Moreover, we find that much work is still needed, in particular when it comes to combining privacy, transparency, scalability, and ease of deployment. We report on over 90 relevant articles and discuss and compare in detail 30 reports on sensing pedestrian behavior.

A Peer-to-Peer Swarm Creation and Management Framework

The continuous development of Internet bandwidth, communication and content variety have enabled peer-to-peer protocols to take place among the most heavily used protocols in the Internet, with BitTorrent as the current flagship. Recent research has focused on measuring Peer-to-Peer systems and proposing improvements and enhancements to existing protocol with the goal of ensuring rapid content distribution and general user satisfaction. In this paper we present a framework that allows easy swarm creation and control for different BitTorrent clients. With the help of a virtualized infrastructure and a client-server software layer we are able to create, command and manage large sized BitTorrent swarms. The framework allows a user to run, schedule, start, stop clients within a swarm and collect information regarding their behavior.