In recent years there has been considerably interest in positioning system for vacuum environment. In general, observation and handling of atomic or nanometer-sized structures are carried out in vacuum environment. Furthermore, optical industry and semiconductor process, a vacuum environment is very important. For the positioning in vacuum environment, advanced actuators are necessary.
Conventional electromagnetic motors cannot be used in vacuum, because these motors generate outgas and cannot withstand without lubricating. In addition, electromagnetic noise might have an adverse influence on optical industry and semiconductor process. Compared with conventional electromagnetic motors, ultrasonic motors have many advantages, such as a large output force with low speed, brakeless mechanism, simple structure, quick response, and electromagnetic immunity etc. Especially, since ultrasonic motor don’t need the lubrication, the ultrasonic motor is one of the promising actuator in vacuum environment. A bolt-clamped Langevin-type transducer (BLT) has been developed to produce large vibration amplitude and large mechanical force. In our laboratory, a BLT is developed with nanometer level accuracy in various atmosphere environments.
The positioning systems using a ball-screw driven stage generally need a controller to satisfy the requirements; high accuracy, fast response, little overshoot, and robustness, because non-linearity and uncertainty of friction disturb the required performance. Many kinds of control methods already proposed and evaluated such as time optimal control, sliding mode control, and disturbance observer. However, these control methods need both a mathematical exact model of plant and parameters. Since, the identification process should be iterative task, several kinds of sensor are needed for the system identification of the plant, Identification of parameter and modeling is time and money consuming process.
To overcome these problems, auto-tuning controllers and intelligent controllers, for instance fuzzy logic control and neural control, have been proposed. However, these control methods tend to be complex and hard to understand their design algorithms. In order to solve these problems, we propose the point-to-point control theory that has no friction model, has easy system identification process and is easy to use in the industry area. The proposed switching control method consists of a macro controller and a micro controller. The macro controller makes the stage to move near the object position. Then, the macro controller is switched to the micro controller. The micro controller is possible to make a stage system positioned at the object point within sub-nanometer error, because the contact surface friction characteristics of a ball-screw driven stage work as a elastic element in the micro dynamics range.