2018 IEEE International Instrumentation
& Measurement Technology Conference
Houston, Texas, USA
- Francesco Santoni, Alessio De Angelis, Antonio Moschitta and Paolo Carbone
Calibrating a magnetic positioning system using a robotic arm
Abstract: A robotic arm moving its end-effector along userdefined trajectories is used to calibrate a Magnetic Positioning System (MPS). The principle of operation of the positioning system is reviewed. Position and attitude of an active coil are estimated measuring the induced voltage on a set of fixed coils, with known position and versor. The Unscented Kalman Filter (UKF) is used to smooth the measured trajectories. In order to estimate the precision up to which ground-truth trajectories are
traced by the robotic arm, a calibration procedure of the robot is illustrated. The active coil is fixed on the end-effector of the robot, and ground-truth trajectories are then tracked in order to calibrate the whole MPS plus UKF system. A wide groundtruth trajectory is used to obtain a fine estimate of positions and versors of the fixed coils. Using the parameters thus obtained, several trajectories are tracked. The UKF is tuned in order to minimize the error affecting the estimated position and attitude of the active coil at each point of any trajectory.
Video shown during the presentation:
Real-time acquisition with Magnetic Positioning System
- Alessio De Angelis, Paolo Carbone and Johan Schoukens
Generating Low-Distortion Ternary Signals Using an Imperfect DAC and ADC Chain
Abstract: In this paper, the problem of generating ternary signals with high spectral purity using imperfect equipment is addressed. In particular, a low-cost setup comprised of a nonuniform digital-to-analog converter (DAC), an analog filter, and a nonuniform analog-to-digital converter (ADC) is considered.
First, the effect of DAC nonlinearity on the spectrum of randomized constrained sequences (RCS) is described and validated. Then, based on this effect, a method for generating RCS signals with high spectral purity is proposed. The feasibility of this method is evaluated by numerical simulations and experimental tests.
- Guido De Angelis, Alessio De Angelis, Antonio Moschitta and Paolo Carbone
Identification of Resonant Circuits’ Parameters Using Weighted-Least-Squares Fitting
Abstract: Identification of RLC circuital parameters is investigated. We present a method for estimating the parameters of the circuit transfer function. The theoretical background is discussed, theoretical models are presented and experimentally validated. A practical implementation is presented and employed to validate the considered numerical models. Numerical and experimental results demonstrate that the proposed system provides a good accuracy measurement of the parameters of the resonant circuit.
The experimental results validate the considered numerical models with a discrepancy of less than 10% with respect to measurement models.