Section: New Results

Applications

• The problem of avoiding obstacles while navigating within an environment for a Unicycle-like Wheeled Mobile Robot (WMR) is of prime importance in robotics. The work [32] solves such a problem proposing a perturbed version of the standard kinematic model able to compensate for the neglected dynamics of the robot. The effectiveness of the solution is proved, supported by experiments and finally compared with the Dynamic Window Approach (DWA) to show how the proposed method can perform better than standard methods. The paper [60] presents a decentralized solution to control a leader-follower formation of unicycle wheeled mobile robots allowing collision and obstacle avoidance. The work [62] solves the obstacle avoidance problem extending the Potential Field (PF) method for a mobile robot. The usual definition of the PF has been modified to have a field which is continuous everywhere. It is shown that the system has an attracting equilibrium at the target point, repelling equilibriums in the centers of the obstacles and saddle points on the borders. Those unstable equilibriums are avoided capitalizing on the established Input-to-State Stability (ISS) property of this multi-stable system. To escape a local minima this work makes the most of ISS property that is not lost for perturbations. And for small properly designed disturbances the global attractivity of the target point is proved.

• The paper [63] investigates the behavior of central Jacobi differentiator in robot identification applications. It is applied to compute acceleration from noisy position measurements. Its frequency domain property is analyzed via a finite impulse response (FIR) filter point of view, indicating clearly the differentiators performance. Two revolute joints planar robot parameter identification is done. Comparisons between the Jacobi differentiator and the Euler differentiation combined with Butterworth filter are drawn.

• In [50] the velocity of valve movement activity is estimated using three different differentiation schemes: an algebraic-based differentiator method, a non-homogeneous higher order sliding mode differentiator and a homogeneous finite-time differentiator. We demonstrate that this estimated velocity can be used for water quality monitoring as the differentiators can detect very rapid change in valve movements of the oyster population resulting from some external stimulus or common input.

• In the paper [15] the measurements of valve activity in a population of bivalves under natural environmental conditions (16 oysters in the Bay of Arcachon, France) are used for a physiological model identification. A nonlinear auto-regressive exogenous (NARX) model is designed and tested. Through this study, it is demonstrated that the developed dynamical model of the oyster valve movement can be used for estimating normal physiological rhythms of permanently immersed oysters and can be considered for detecting perturbations of these rhythms due to changes in the water quality, i.e. for ecological monitoring.

• Spawning observations are important in aquaculture and biological studies, and until now, such a detection is done through visual analysis by an expert. Using measurements of valve activity (i.e. the distance between the two valves) in populations of bivalves under natural environmental condition (16 oysters in the Bay of Arcachon, France, in 2007, 2013 and 2014), algorithms for an automatic detection of the spawning period of oysters are proposed in the paper [16] , [51] . The fault detection method presented in the paper can also be used to detect complex oscillatory behavior which is of interest to control engineering community.

• The work presented in the paper [33] is undertaken within the European FP7 funded Advanced Fault Diagnosis for Sustainable Flight Guidance and Control (ADDSAFE) project. It proposes new fault detection and fault diagnosis techniques that could significantly help developing environmentally-friendlier aircraft. LPV model-based fault detection schemes are proposed and compared for robust and early detection of faults in aircraft control surfaces servo-loop. The proposed methodologies are based on a slight modification of the $H_{\infty }$∕$H_{-}$ LPV optimization techniques for systems modelled in, first polytopic manner, second linear fractional representation fashion. It is shown that the proposed fault detection schemes can be embedded within the structure of in-service monitoring systems as a part of the Flight Control Computer software. Several important examples on model and signal based fault detection in aircraft Electrical Flight Control System are studied in [80] .

• For analyzing the transients of induction heating systems, time-dependent phasor transformations were proposed so far in the literature. Applying these transformations to a linear R, L, C circuit equations leads to differential equations in the complex domain from which equivalent circuits modeling the envelopes of sinusoidal waveforms were derived. The work [46] proposes a phasor transformation which is based on fictitiously replacing the real voltage and current signals of a system by complex ones. It leads to transformed system equations in the real domain where instantaneous amplitudes, phases and frequencies appear explicitly, which makes the transformed equations suitable for the feedback control design. The methodology is applied to a parallel induction heating system in order to design a sliding mode controller.

• The problem of air-to-fuel ratio regulation for a direct injection engine is addressed in [54] . A LPV model of the engine is used, for which an interval observer is designed. The interval observer is applied for the model validation and control synthesis. The results of design are confirmed by implementation.

• Modular Robot Manipulators are user-configurable manipulators which provide rapid design and inexpensive implementation. To be easy-use, smart actuators embedded with position input and position feedback controller are adopted, these local controllers render the manipulators position controlled, but also result in limited performance and precision. The paper [72] targets the case that the built-in controller does not provide desirable precision for set-point regulation. Firstly a joint-level model is established, of which the nominal model can be identified with derivative observer based on the position feedback, then an auxiliary adaptive controller coping with parametric uncertainty is proposed which leads to an error close to zero, a switching control strategy is introduced considering the actuator saturation. The paper [73] addresses the set-point control of actuators integrated with built-in controller, which presents steady-state error (SSE) under certain load. To eliminate the SSE, a model of the actuator-plus-controller system is established and identified, a switched adaptive controller is developed to work with the embedded one, considering the physical constraints, a switching control strategy is proposed. The proposed algorithms are implemented on a 5-DOF modular manipulator, with comparison to classic integral controller.

• The communication [74] is devoted to a comparison between various meteorological forecasts, for the purpose of energy management, via different time series techniques. The first group of methods necessitates a large number of historical data. The second one does not and is much easier to implement, although its performances are today only slightly inferior. Theoretical justifications are related to methods stemming from a new approach to time series, artificial neural networks, computational intelligence and machine learning.

• ALINEA is a well known ramp metering closed-loop control the aim of which is to improve highway traffic. The report [84] shows that ALINEA may be slightly modified in order to be efficiently implemented without any need of crucial time-varying quantities, like the critical density and the free-flow speed, which are most difficult to estimate correctly online.

• For malaria patients, a usual observation problem consists in estimation of sequestered parasites Plasmodium falciparum from measurements of circulating ones. The model of an infected patient is rather uncertain, and for all rates (death, transition, recruitment and infection) in the model it is assumed that only intervals of admissible values are given. In addition, the measurements of the concentration of circulating parasites are subjected by a bounded noise, while some parameters, like the rate of infection of blood cells by merozoites, are completely unknown and highly time-varying. In order to evaluate the concentration of sequestered parasites, an interval observer is designed in [85] , which provides intervals of admissible value for that concentration, with the interval width proportional to the model uncertainty.