Robust, flexible position control of mobile robots
Microsystems and nano-technologies undergo rapid development and find key applications in numerous fields such as automotive, medicine, biotechnology, telecommunication and consumer electronics to name a few. One of the major parameters for the wider expansion of these technologies is the availability of suitable operating tools. Answering this need the MICRON project resulted into a prototype of a multi-microrobot manipulating system for handling of micrometre-sized and mesoscopic objects with nanometre precision. This system is a cluster of small mobile robots that co-operate autonomously for accomplishing delicate tasks such as the handling of biological cells and the assembly of micro-parts. With focus on the development of the prototype system, the project resulted into several product and process innovations, including a robust and flexible position control system. The high complexity of tasks to be performed requires effective and adaptable control of the dynamic behaviour of the robots used. For this reason, several issues were taken into consideration including the completely different movement behaviours of various robots employed in the system. The size and the drive command such as number of mechanical degrees of freedom may also differ among robots. The high complexity of the assigned tasks can easily lead to changing conditions and variant parameters in time. Moreover, the non-linearity of the robotic movement could result into a chaotic situation at high velocities for piezo-driven robots. An additional constraint involves the prevention of the simultaneous work of two closed loops of control. This is mainly due the Action-based Environment Modelling (AEM)-based navigation employed that depends on the operation mode. Taking into account all these issues the MICRON project adopted a two-staged approach for the reliable position control. This consists of a driver working in open loop to equilibrate non-linearities and balance the differences between different robots. At a higher level, a closed loop controller based on an almost linearised driver compensates for the remaining deviation. For further information click at: http://wwwipr.ira.uka.de/~micron/index.html(odnośnik otworzy się w nowym oknie)
