Michele Ermidoro

Development of a wireless low-power multi-sensor network for motion tracking applications

This work presents a novel wireless and low power Attitude and Heading Reference Systems network based on low-cost MEMS (Micro Electro-Mechanical System) sensors, developed for motion tracking systems. Biomedical and rehabilitation purposes as well as gaming and consumer electronics may be the potential applications of this network. The paper aims to describe the hardware architecture, the embedded sensor fusion algorithm and the motion tracking system.

On time-optimal anti-sway controller design for bridge cranes

In this paper, we analyze and discuss the time optimal control problem for bridge cranes. Specifically, we employ the geometric approach introduced in [12] to address the control issue without computing the analytic trajectory of the costate. As a second contribution, we show that an alternative approach based on a proportional controller accounting for the limits of the control action can be tuned to resemble to the time-optimal solution, but with simpler and more robust implementation. The above ideas are validated by means of some simulation examples.

Development of an Attitude and Heading Reference System for Motion Tracking Applications

This work presents a novel wireless and low-power platform based on low-cost MEMS (microelectromechanical system) sensors to be used for motion tracking and navigation systems. Biomedical and rehabilitation purposes as well as gaming and consumer electronics may be the potential applications. The paper aims to describe the hardware architecture, the embedded sensor fusion algorithm providing real-time orientation of the platform, and the performance with respect to the state-of-the-art solutions.

Anti-sway fixed-order control of bridge cranes with varying rope length

In this paper, a gain-scheduling control law is proposed to attenuate the oscillations in overhead cranes induced by manual operations, for different values of the rope length. To take into account the practical limits in controller implementation, a fixed-order controller is tuned, by enforcing certain robustness and performance constraints. The proposed strategy is experimentally tested on a real bridge crane and compared to a time-invariant solution.

Load swing reduction in manually operated bridge cranes

In this paper, a procedure for control system design of an anti-sway system for manually operated bridge cranes is presented. Specifically, the human operator acts by means of a push-button panel to operate the crane and thus is considered as a disturbance to the closed loop control system that regulates to zero the angle of the load cable with respect to the vertical axis. Closed loop control has been made possible by the use of an inertial sensor and an Extended Kalman Filter which provides accurate angle estimation. An accurate multi-body model has been realized using the Automatic Dynamic Analysis of Mechanical System (ADAMS) environment. The model parameters have been tuned by system identification based on real measurements. Finally, the model is used to design a Sliding Mode Controller to suppress the load swing angle.

A condition assessment algorithm based on dynamic time warping for high-voltage circuit breaker

The paper presents an innovative algorithm for the analysis of the residual life of an high-voltage circuit breaker. This algorithm relies on the measurements of the accelerometer and on the Dynamic Time Warping algorithm (DTW). The DTW is used to analyze the spectrogram of the acceleration measured, permitting an analysis both in time and frequency. The indexes created are then compared to the values registered during the initial phase, firing an alarm or a simple warning depending on the deviation. The algorithm is tested on two different circuit breaker, showing its effectiveness in predicting failures.