In future many elderly people will have to stay longer in their homes. This requires intelligent monitoring of the elderly. When somebody falls, does not stand up or requires help, ambient intelligence in the house should detect accidents or health emergencies and take action like warning a care center. Besides the ambient intelligence in the house also a mobile interaction system is required with direct sensing of the person observed. The robot companion helps a person with the communication with the intelligent house and the communication with a care center when needed. This project is concerned with the development of such a mobile personal robot companion.
This team has to:
Develop a simulation environment in which a map is shown and on it the location of a number of sensors. An XML script simulates sensor values that also need to be shown on the map. In addition to the simulated sensors some real sensors are used in the simulator.
This team has to:
- Review 'Java Camera' paper
- Review 'Probabilistic Resource Allocation in Distributed Fusion Systems' paper
- Review 'Interactive response system for crisis management' paper
- Review 'Intelligent Traffic Light Control' paper
An intelligent camera, programmable in Java must be extended with possibilities to calculate the direction of movement and to allow counting of the number of people entering a certain area. Other application areas may also be considered.
This team has to:
This project is focusing on resource allocation in Distributed Perception Networks (DPN), an agent-based approach to the fusion of large amounts of heterogeneous information. Based on a mathematical analysis the team will develop and implement an efficient approach for the resource allocation based on the evidence impact on hypothesis nodes in the perception networks. The resulting resource allocation algorithms will enhance the existing DPN toolbox.
This team has to:
After a disaster occurs (imagine an explosion of a chemical plant, or a derailed train) usually choas results. Soon after that, crisis fighting teams (police, firesquad and medical) will rush into the crisis area. These teams are instructed by a control room that organizes their activities. The control room receives information about the ongoing crisis through several sources. One very important source of information is human observations. Humans that are near the crisis area will start calling 112 by mobile phone, and will most likely continue to report what they see.
At the UvA, it is studied how the functionality of the control room can be improved such that better crisis management will result. To improve crisis management, it is important that much information about the crisis is available. In particular, support from the humans that observe the crisis is studied. Imagine a system that would call and ask those people very directed questions (such as: "do you see fire" or "do you see casualties?"). The answers are then used by the system to update its perception about the crisis situation. The results are used by the control room operators to better manage the crisis.
This project focuses on the development of such an interactive crisis management system. The intended work is both practical and theoretical. The practical work includes the development of an interactive phone-system using SMS messages. The theoretical part will focus on the situation analysis of a (fictive) crisis. Previous work on this topic by the UvA can be used to quickly achieve interesting results.
This team has to:
A recently emerging theme within AI and related disciplines is the control of complex, distributed, multi-agent systems. Examples of such systems include teams of robotic soccer players, sensor networks, and routers for internet or cell phone data. In this project, the focus will be on road traffic management. Using an existing traffic simulator, a variation of an intelligent, learning traffic light controller will be implemented and compared to other learning controllers and to standard controllers.
This team has to:
This year will be the 8th edition of the RoboCupSoccer - 4 Legged League. The technical committee of this League has decided to remove the walls around the playing field of the robotdogs. This will have major impact on the algorithms to control the Aibo's. It is the task of the Masterstudents to design and evaluate algorithms that can cope with a soccerfield marked out just by lines.
This team has to: