Autonomous emergent painting triggered by music and motion

Multi-agent systems can be considered as a part of distributed artificial intelligence approaches. They are particularly characterised by emerging global properties which stem from combinations of local interaction between individual pro-active elements called agents. Multi-agent systems have already been used in a wide range of applications but few applications integrate interaction with external devices and/or human beings. This paper presents a multi-agent-based application aimed, at producing emergent paintings based on both internal classical interaction and external triggering devices, such as music/sounds sources and motion captured by accelerometer sensors. The application results are computer generated paintings which can be classified as cyberart. The main challenge of this work is to obtain pertinent emergent paintings in relation with external trigger specificities.