Robotic systems are a notable example of the class of reactive systems in which design complexity faces not only timing constraints and correct ordering in execution, but also spatial relationships and dynamic evolution of robotic units. Simulation of agents in a 3D environment is a way for getting a better understanding of system behavior as well as detecting possible hazardous conditions concerning event sequencing and timing, along with their relations with spatial constraints.

In this paper, a formal high-level control model based on Time Petri Nets is used to capture robot behavior and to drive animation in the virtual 3D environment. A rich set of tools has been developed to allow assembly of virtual robots, their composition into a virtual workcell and the definition of the control model itself.