Multi-Agent Systems

Here are some recent research projects that are related to my research focus in the Multi-Agent Systems area. My research is based on smart agents that act in real-time, dynamic, uncertain, and adversarial environments. Some research projects have simulated agents, such as avatars in serious games (e.g. the work with Dr. Lisetti from FIU). Other simulated agents are agents in games, such as soccer agents in the awarded soccer manager (see below). A big part of my research are also pysical agents, namely NAO robots. Here are some examples:

RoboViz is a software program designed to assess and develop agent behaviors in a multi-agent system, the RoboCup 3D simulated soccer league. RoboViz is an interactive monitor that renders agent and world state information in a three-dimensional scene. In addition, RoboViz provides programmable drawing and debug functionality to agents that can communicate over a network.
More details here.
RoboCanes walking engine. A new walking engine for adaptive walking on bipedal robots. It is a LIPM-based closed-loop walk, that adapts to differences in the physical behavior of the robot by optimizing parameters of the model directly on the NAO while walking and executing other tasks. A significant amount of errors in the model predictions can be reduced.
See video here.
  MoCap. Our motion capture lab is based on passive markers with retroreflective material (Vicon system). There are 10 4-megapixel cameras (120 Hz) and 4 force plates to measure forces on ground. There are also facilities to capture facial expressions and electromagnetic devices to capture voltage in max 16 muscles groups for biomedical studies. We made experiments with complex sports motions using soccer as an example. The 3D data are then ported to a 3D animated soccer player in a video game. See video here.
Automatic plan and intention recognition of opponent mobile robots in adversarial and dynamic environments. Project supported by DFG SPP 1125 (German Research Council) from 1999 - 2005. 16 publications covering a range of topics such as behavioral architectures of MAS, spatial knowledge representation and reasoning, online leaning, and vision-based fast and reactive Monte-Carlo localization. Robots: Soccer Simulation (Virtual Werder) and Sony AIBO's (German Team). More details here.