Interference checking approach with tolerance based on assembly dimension chain

CAD model with nominal dimension is implemented in interference checking of assembly simulation of aircraft complex parts at present, which causes inadequate availability. In order to address this challenging issue, interference checking method with tolerance based on assembly dimension chain was proposed. Worst case and maximum error probability of tolerance of composing loop were used, and CAD models were respectively re-constructed and inserted into simulation system. Before dynamic interference checking, engineering semantic interference condition was set to assembly requirements. Finally, the interface checking result was a basis for reasonability of assembly process and tolerance. A prototype system was developed based on the above research.

A hand gesture-based interactive model to choose one from many

Natural, efficient, intelligent, and barrier-free HCI （Human-Computer Interface） will be the main developing direction of the new intelligent HCI. Building an intelligent, natural, harmonious and humanistic HCI has been the primary developing tendency of the new HCI. ＂Select one from many＂ is one of the key questions in researching and designing HCI based on gestures. On the sin- gle channel interactive condition which is based on 3D gesture operation, we build a interactive model of＂Select one from many＂. Combining this model with ＂Token＂ technology and reciprocating movement feature, we present and realize a simple efficient ＂Se- lect one from many＂ algorithm, which is used in a concrete SOR （Selection-Operation-Release） system. In the SOR system, we real- ize a real-time selection, tracking and interaction of natural gestures.

Crucial problems in the design of a terahertz tripler

A frequency-multiplied source at the terahertz band using discrete planar Schottky diodes, which is a critical element in heterodyne instruments, has been studied by some domestic research institutions in recent years. Besides the design method, there are still many crucial problems that must be taken into consideration in the design. This article mainly discuss three aspects based on the measured data of a 225 GHz tripler that we designed. Firstly, the accuracy of the diode model concerns the reliability of the simulation results. According to the Spice parameters and the measured results, the physical size and the DC parameter of the Schottky diode can be corrected until there is a good consistency between the simulated and measured results. Secondly, the heat accumulation happens to the Schottky junction when the high input power is added. A steady-state thermal simulation is done and the results show that the hottest temperature is about 140℃ with 250 mW input power, which is safe to the diode. Lastly, some non-ideal factors are brought during the assembly process such as the uncertainty in the conductive adhesive shape and location deviation of the circuit. Furthermore, the effect on the performance of the frequency multiplier is calculated in this work.

Calibration of robotic drilling systems with a moving rail

Industrial robots are widely used in aircraft assembly systems such as robotic drilling systems. It is necessary to expand a robot＇s working range with a moving rail. A method for improving the position accuracy of an automated assembly system with an industrial robot mounted on a moving rail is proposed. A multi-station method is used to control the robot in this study. The robot only works at stations which are certain positions defined on the moving rail. The calibration of the robot system is composed by the calibration of the robot and the calibration of the stations.The calibration of the robot is based on error similarity and inverse distance weighted interpolation.The calibration of the stations is based on a magnetic strip and a magnetic sensor. Validation tests were performed in this study, which showed that the accuracy of the robot system gained significant improvement using the proposed method. The absolute position errors were reduced by about 85%to less than 0.3 mm compared with the maximum nearly 2 mm before calibration.