A Real-Time Optical Sensor Simulation Framework for Development and Testing of Industrial and Mobile Robot Applications

Abstract

Sensors are an integral part of most robot applications today. Therefore, sensor simulation is an important aspect in many simulation scenarios dealing with robotics. This contribution introduces new methods for optical sensor simulations such as (stereo-) cameras, Time-of-Flight (TOF) cameras or laser range scanners. The algorithms use modern programmable graphics hardware for a close-to-reality simulation of the sensors in real-time - even for complex application scenarios. They are integrated in a flexible sensor simulation framework capable of controlling and recording as well as simulating a wide range of sensors. This framework is a key component of our Virtual Testbed approach providing a versatile engineering environment for the development and test of robotic applications. The practical applicability is demonstrated in different scenarios, ranging from industrial robot applications to the virtual prototyping of robots for space exploration.