Nicolae Jula

Inertial sensor signals denoising with directed transfer function

In order to achieve precise navigation information on the displacement of a vehicle in terms of position, speed, acceleration and direction we are suggesting the use of a new software procedure for processing data received from miniaturized inertial navigation systems. A new method for establishing the optimum decomposition level of the Wavelet transform is proposed. This method was used for filtering variable acceleration signals received from an accelerometer. In our next investigations we intend to implement this algorithm for the pre-calibration of two or tri-dimensional navigation systems containing miniaturized accelerometers and gyros.

Assistive inertial 3D positioning MEMS system

This paper describes the implementation and experimental validation of a new 3D miniaturized inertial navigator developed for human indoor and outdoor monitoring applications. Accuracy is a major concern, especially during GPS outages and indoor applications. In order to increase the inertial navigation system (INS) standalone operation period with acceptable accuracy during the GPS absence, signals received from sensors are filtered for reducing positioning errors. For validation, data achieved from the proposed navigator was compared with data achieved from a GPS/INS integrated navigator, both devices being used as remote systems. Unlike in our previous research, the case of the bi-dimensional model, when we managed to develop an optimizing tuning method for achieving the optimum level of decomposition of the Wavelet filter for data achieved from two channels, the attitude channel and the position channel, but not simultaneously, we are currently extending our research aiming to tune simultaneously data achieved from six detection channels, three accelerometers and three gyros.

OPTIMIZATION OF A SMART ACCELEROMETER BASED ON AMPLITUDE- FREQUENCY CHARACTERISTICS CONSTRAINTS

The paper deals with a smart inertial sensor architecture optimization by using a tuning algorithm related to some constraints imposed to the sensor amplitude-frequency characteristics. To test the optimization criterion the model of a close loop accelerometer is used. Starting from the basic architecture of the sensor, based on classical control at the loop closing, a new smart architecture is proposed. In this way, a fuzzy logic controller is added on the direct path of the accelerometer in order to replace an amplification and filtering block used initially to create the control the feedback force acting on the proof mass. Further, a Matlab/Simulink model is developed for accelerometer and a method to plot its amplitudefrequency characteristics based on this model is described. An optimization subroutine for the tuning of derivative gain of the fuzzy logic controller is presented and tested on the simulation model.

Techniques of aquisition, processing and validation of satellite images

Within the European Politics for Geographic Information, geographic information represents a complex part in continuous development of informational society having a wide range of applicability. This can be categorized into: spatial, geographic information and descriptive, qualitative information. These two types of data can be integrated in the same information management system by a Geographic Informational System. (GIS).