Control of dynamical systems with temporal logic specifications

This thesis is motivated by safety-critical applications involving autonomous air, ground, and space vehicles carrying out complex tasks in uncertain and adversarial environments. We use temporal logic as a language to formally specify complex tasks and system properties. Temporal logic specifications generalize the classical notions of stability and reachability that are studied in the control and hybrid systems communities. Given a system model and a formal task specification, the goal is to automatically synthesize a control policy for the system that ensures that the system satisfies the specification. This thesis presents novel control policy synthesis algorithms for optimal and robust control of dynamical systems with temporal logic specifications. Furthermore, it introduces algorithms that are efficient and extend to high-dimensional dynamical systems.

The first contribution of this thesis is the generalization of a classical linear temporal logic (LTL) control synthesis approach to optimal and robust control. We show how we can extend automata-based synthesis techniques for discrete abstractions of dynamical systems to create optimal and robust controllers that are guaranteed to satisfy an LTL specification. Such optimal and robust controllers can be computed at little extra computational cost compared to computing a feasible controller.

The second contribution of this thesis addresses the scalability of control synthesis with LTL specifications. A major limitation of the standard automaton-based approach for control with LTL specifications is that the automaton might be doubly-exponential in the size of the LTL specification. We introduce a fragment of LTL for which one can compute feasible control policies in time polynomial in the size of the system and specification. Additionally, we show how to compute optimal control policies for a variety of cost functions, and identify interesting cases when this can be done in polynomial time. These techniques are particularly relevant for online control, as one can guarantee that a feasible solution can be found quickly, and then iteratively improve on the quality as time permits.

The final contribution of this thesis is a set of algorithms for computing feasible trajectories for high-dimensional, nonlinear systems with LTL specifications. These algorithms avoid a potentially computationally-expensive process of computing a discrete abstraction, and instead compute directly on the system's continuous state space. The first method uses an automaton representing the specification to directly encode a series of constrained-reachability subproblems, which can be solved in a modular fashion by using standard techniques. The second method encodes an LTL formula as mixed-integer linear programming constraints on the dynamical system. We demonstrate these approaches with numerical experiments on temporal logic motion planning problems with high-dimensional (10+ states) continuous systems.

A dedicated binding mechanism for the visual control of movement.

The human motor system is remarkably proficient in the online control of visually guided movements, adjusting to changes in the visual scene within 100 ms [1-3]. This is achieved through a set of highly automatic processes [4] translating visual information into representations suitable for motor control [5, 6]. For this to be accomplished, visual information pertaining to target and hand need to be identified and linked to the appropriate internal representations during the movement. Meanwhile, other visual information must be filtered out, which is especially demanding in visually cluttered natural environments. If selection of relevant sensory information for online control was achieved by visual attention, its limited capacity [7] would substantially constrain the efficiency of visuomotor feedback control. Here we demonstrate that both exogenously and endogenously cued attention facilitate the processing of visual target information [8], but not of visual hand information. Moreover, distracting visual information is more efficiently filtered out during the extraction of hand compared to target information. Our results therefore suggest the existence of a dedicated visuomotor binding mechanism that links the hand representation in visual and motor systems.

LOCAL DIGITAL CONTROL OF POWER ELECTRONIC CONVERTERS IN A DC MICROGRID BASED ON A-PRIORI DERIVATION OF SWITCHING SURFACES

In power electronic basedmicrogrids, the computational requirements needed to implement an optimized online control strategy can be prohibitive. The work presented in this dissertation proposes a generalized method of derivation of geometric manifolds in a dc microgrid that is based on the a-priori computation of the optimal reactions and trajectories for classes of events in a dc microgrid. The proposed states are the stored energies in all the energy storage elements of the dc microgrid and power flowing into them. It is anticipated that calculating a large enough set of dissimilar transient scenarios will also span many scenarios not specifically used to develop the surface. These geometric manifolds will then be used as reference surfaces in any type of controller, such as a sliding mode hysteretic controller. The presence of switched power converters in microgrids involve different control actions for different system events. The control of the switch states of the converters is essential for steady state and transient operations. A digital memory look-up based controller that uses a hysteretic sliding mode control strategy is an effective technique to generate the proper switch states for the converters. An example dcmicrogrid with three dc-dc boost converters and resistive loads is considered for this work. The geometric manifolds are successfully generated for transient events, such as step changes in the loads and the sources. The surfaces corresponding to a specific case of step change in the loads are then used as reference surfaces in an EEPROM for experimentally validating the control strategy. The required switch states corresponding to this specific transient scenario are programmed in the EEPROM as a memory table. This controls the switching of the dc-dc boost converters and drives the system states to the reference manifold. In this work, it is shown that this strategy effectively controls the system for a transient condition such as step changes in the loads for the example case.

Automated symbolic and numeric procedure for manipulator rigid-body dynamic significance analysis and simplification

This paper describes an automated procedure for analysing the significance of each of the many terms in the equations of motion for a serial-link robot manipulator. Significance analysis provides insight into the rigid-body dynamic effects that are significant locally or globally in the manipulator's state space. Deleting those terms that do not contribute significantly to the total joint torque can greatly reduce the computational burden for online control, and a Monte-Carlo style simulation is used to investigate the errors thus introduced. The procedures described are a hybrid of symbolic and numeric techniques, and can be readily implemented using standard computer algebra packages.

MWCNT操作的动力学运动学建模研究

为实现MWCNT操作运动过程的视觉显示,本文建立了MWCNT的动力学模型,据此可推导出推动MWCNT所需施加力的大小,并根据探针的实际受力判断其能否运动;同时还建立了MWCNT的运动学模型,根据探针的实际位置可获得探针操作下MWCNT的新位置与姿态,并借助虚拟现实技术对视觉界面进行实时更新,实现了MWCNT运动过程的实时视觉显示。基于上述视觉显示,操作者可在线控制探针的作用位置与运动轨迹、以及施加在探针上作用力的大小与方向,实现对MWCNT操作过程及结果的在线控制。MWCNT的操作实验初步验证了该模型的有效性。

A comparison of real catching with catching using stereoscopic visual displays

To date, the usefulness of stereoscopic visual displays in research on manual interceptive actions has never been examined. In this study, we compared the catching movements of 8 right-handed participants (6 men, 2 women) in a real environment (with suspended balls swinging past the participant, requiring lateral hand movements for interception) with those in a situation in which similar virtual ball trajectories were displayed stereoscopically in a virtual reality system (Cave Automated Virtual Environment [CAVE]; Cruz-Neira, Sandin, DeFranti, Kenyon, & Hart, 1992) with the head fixated. Catching the virtual ball involved grasping a lightweight ball attached to the palm of the hand. The results showed that, compared to real catching, hand movements in the CAVE were (a) initiated later, (b) less accurate, (c) smoother, and (d) aimed more directly at the interception point. Although the latter 3 observations might be attributable to the delayed movement initiation observed in the CAVE, this delayed initiation might have resulted from the use of visual displays. This suggests that stereoscopic visual displays such as present in many virtual reality systems should be used circumspectly in the experimental study of catching and should be used only to address research questions requiring no detailed analysis of the information-based online control of the catching movements.

Étude du caractère automatique du processus de contrôle en ligne lors de tâche de pointage manuel

Lors d’une tâche de pointage manuel, la présence de corrections rapides, adaptées, automatiques et même réflexes (Franklin et Wolpert, 2008) suite à une perturbation par saut de curseur a pu être observée dans de nombreuses études. Ici, nous avons souhaité déterminer si ces corrections étaient purement réflexes où si elles étaient amorcées seulement lorsque la perturbation mettait en péril l’atteinte de la cible ; ces corrections ont-elles aussi un aspect fonctionnel ? Dans une première expérience nous avons fait varier la taille des cibles (5 ou 30 mm de diamètre) et des sauts du curseur (5, 15 ou 25 mm) de manière à obtenir certaines combinaisons où la cible pourrait être atteinte sans qu’aucune correction du mouvement pour contrecarrer l’effet du saut du curseur ne soit nécessaire. Des corrections réduisant l’erreur d’environ 65% ont été observées dans toutes les conditions. Dans une seconde expérience, les participants devaient atteindre une très grande cible (arc de 30°) et un saut de curseur de 15 mm était introduit pour certains essais peu de temps après l’amorce du mouvement. Les participants ont modifié leur mouvement dans le sens opposé à celui de la perturbation, et cela même s’ils n’avaient pas détecté consciemment le saut. Cependant, ces corrections étaient moins rapides et plus petites (42% de l’amplitude du saut de curseur) que celles observées lors de la première expérience. Nos résultats supportent le fait que l’amorce des corrections pour des erreurs de trajectoire induites expérimentalement soit de nature réflexe. Un deuxième processus serait alors responsable du déroulement de ces corrections ; ce deuxième processus est basé, entre autres, sur les caractéristiques de la cible.

L’influence d’un contexte virtuel sur les processus de contrôle en ligne des mouvements d’atteinte manuelle

L’objectif principal de la présente thèse était de déterminer les facteurs susceptibles d’influencer l’efficacité des processus de contrôle en ligne des mouvements d’atteinte manuelle. De nos jours, les mouvements d’atteinte manuelle réalisés dans un environnement virtuel (déplacer une souris d’ordinateur pour contrôler un curseur à l’écran, par exemple) sont devenus chose commune. Par comparaison aux mouvements réalisés en contexte naturel (appuyer sur le bouton de mise en marche de l’ordinateur), ceux réalisés en contexte virtuel imposent au système nerveux central des contraintes importantes parce que l’information visuelle et proprioceptive définissant la position de l’effecteur n’est pas parfaitement congruente. Par conséquent, la présente thèse s’articule autour des effets d’un contexte virtuel sur le contrôle des mouvements d’atteinte manuelle. Dans notre premier article, nous avons tenté de déterminer si des facteurs tels que (a) la quantité de pratique, (b) l’orientation du montage virtuel (aligné vs. non-aligné) ou encore (c) l’alternance d’un essai réalisé avec et sans la vision de l’effecteur pouvaient augmenter l’efficacité des processus de contrôle en ligne de mouvement réalisés en contexte virtuel. Ces facteurs n’ont pas influencé l’efficacité des processus de contrôle de mouvements réalisés en contexte virtuel, suggérant qu’il est difficile d’optimiser le contrôle des mouvements d’atteinte manuelle lorsque ceux-ci sont réalisés dans un contexte virtuel. L’un des résultats les plus surprenants de cette étude est que nous n’avons pas rapporté d’effet concernant l’orientation de l’écran sur la performance des participants, ce qui était en contradiction avec la littérature existante sur ce sujet. L’article 2 avait pour but de pousser plus en avant notre compréhension du contrôle du mouvement réalisé en contexte virtuel et naturel. Dans le deuxième article, nous avons mis en évidence les effets néfastes d’un contexte virtuel sur le contrôle en ligne des mouvements d’atteinte manuelle. Plus précisément, nous avons observé que l’utilisation d’un montage non-aligné (écran vertical/mouvement sur un plan horizontal) pour présenter l’information visuelle résultait en une importante diminution de la performance comparativement à un montage virtuel aligné et un montage naturel. Nous avons aussi observé une diminution de la performance lorsque les mouvements étaient réalisés dans un contexte virtuel aligné comparativement à un contexte naturel. La diminution de la performance notée dans les deux conditions virtuelles s’expliquait largement par une réduction de l’efficacité des processus de contrôle en ligne. Nous avons donc suggéré que l’utilisation d’une représentation virtuelle de la main introduisait de l’incertitude relative à sa position dans l’espace. Dans l’article 3, nous avons donc voulu déterminer l’origine de cette incertitude. Dans ce troisième article, deux hypothèses étaient à l’étude. La première suggérait que l’augmentation de l’incertitude rapportée dans le contexte virtuel de la précédente étude était due à une perte d’information visuelle relative à la configuration du bras. La seconde suggérait plutôt que l’incertitude provenait de l’information visuelle et proprioceptive qui n’est pas parfaitement congruente dans un contexte virtuel comparativement à un contexte naturel (le curseur n’est pas directement aligné avec le bout du doigt, par exemple). Les données n’ont pas supporté notre première hypothèse. Plutôt, il semble que l’incertitude soit causée par la dissociation de l’information visuelle et proprioceptive. Nous avons aussi démontré que l’information relative à la position de la main disponible sur la base de départ influence largement les processus de contrôle en ligne, même lorsque la vision de l’effecteur est disponible durant le mouvement. Ce résultat suggère que des boucles de feedback interne utilisent cette information afin de moduler le mouvement en cours d’exécution.

Calibration of closed loop controllers for setting impedances in force-reflecting systems

This paper discusses a new method of impedance control that has been successfully implemented on the master robot of a teleoperation system. The method involves calibrating the robot to quantify the effect of adjustable controller parameters on the impedances along its different axes. The empirical equations relating end-effector impedance to the controller's feedback gains are obtained by performing system identification tests along individual axes of the robot. With these equations, online control of end-effector stiffness and damping is possible without having to monitor joint torques or solving complex algorithms. Hard contact conditions and compliant interfaces have been effectively demonstrated on a telemanipulation test-bed using appropriate combinations of stiffness and damping settings obtained by this method.

The quiet eye and tasks demands: Do tougher shots need a quieter eye?

The Quiet Eye (QE; Vickers 1996) has been shown to underpin successful performance, differentiating both expertise (inter-individual) and proficiency (intra-individual), with experts and successful attempts characterised by longer QE durations. The QE is proposed to reflect the time needed to organise and fine tune the parameters of movement (e.g. force and direction). In order to examine this prediction and build upon previous research we experimentally manipulated the difficulty of a golf putting task; we hypothesised that if the QE is related to motor programming then a more difficult task should be associated with longer QE durations. 33 golfers (M age= 21.16, SD= 3.98) with an average handicap of 6.5 (SD= 6.02) performed putts in 4 conditions of increasing difficulty. We manipulated the length of the golf putt (short-4ft, long-8ft) and the contact point of the putter head (large-1.7cm, small-0.5cm,) giving increasingly difficult putting conditions of short-large [1], short-small [2], long-large [3] and long-small [4]. We measured performance (radial error from hole in cm) and QE (in ms) for 10 putts in each condition. A repeated measures ANOVA was performed on the performance and QE data. The performance data suggest that we were successful in increasing the difficulty of the task (F (3,93) = 26.46. p = .000), with the best performance in condition [1] (8.57cm), followed by [2] (9.10cm) followed by [3] (16.11cm) and finally the worst performance was in condition [4] (23.40cm). The QE data suggest that, in keeping with our hypothesis, the QE was lengthened as task difficulty increased (F (3,87) = 11.91, p = .043). The QE was shortest in condition [1] (1787.85ms) and increased to condition [2] (1939.78ms) and condition [3] (2076.51ms), with the longest QE in condition [4] (2164.08ms). More detailed analysis of the QE reveal that it was the proportion of the QE that occurred before movement initiation (pre-QE) which increased with shot difficulty, rather than the proportion that occurred during the swing (Online-QE; see Vine et al., 2013). Results support the notion that more complex tasks are associated with a longer QE duration, specifically participants appear to spend longer fixating the target prior to movement. This likely reflects the time needed to process visual information gathered in a pre-performance routine, to inhibit external distraction, and to pre-programme the increasingly difficult parameters of the movement. Vickers, J.N. (1996). Visual control when aiming at a far target. Journal of Experimental Psychology: Human Perception and Performance, 22, 342-354. Vine, S.J. et al. (2013). Quiet eye and choking: Online control breaks down at the point of performance failure. Medicine and Science in Sports and Exercise, 45, 1988-1994.

Commande prédictive non linéaire d’un séchoir à lit fluidisé pour granules pharmaceutiques

Avec la disponibilité de capteurs fiables de teneur en eau exploitant la spectroscopie proche infrarouge (NIR pour near-infrared) et les outils chimiométriques, il est maintenant possible d’appliquer des stratégies de commande en ligne sur plusieurs procédés de séchage dans l’industrie pharmaceutique. Dans cet ouvrage, le séchage de granules pharmaceutiques avec un séchoir à lit fluidisé discontinu (FBD pour fluidized bed dryer) de taille pilote est étudié à l’aide d’un capteur d’humidité spectroscopique. Des modifications électriques sont d’abord effectuées sur le séchoir instrumenté afin d’acheminer les signaux mesurés et manipulés à un périphérique d’acquisition. La conception d’une interface homme-machine permet ensuite de contrôler directement le séchoir à l’aide d’un ordinateur portable. Par la suite, un algorithme de commande prédictive (NMPC pour nonlinear model predictive control), basée sur un modèle phénoménologique consolidé du FBD, est exécuté en boucle sur ce même ordinateur. L’objectif est d’atteindre une consigne précise de teneur en eau en fin de séchage tout en contraignant la température des particules ainsi qu’en diminuant le temps de lot. De plus, la consommation énergétique du FBD est explicitement incluse dans la fonction objectif du NMPC. En comparant à une technique d’opération typique en industrie (principalement en boucle ouverte), il est démontré que le temps de séchage et la consommation énergétique peuvent être efficacement gérés sur le procédé pilote tout en limitant plusieurs problèmes d’opération comme le sous-séchage, le surséchage ou le surchauffage des granules.

The influence of online reviews on consumers’ attributions of service quality and control for service standards in hotels

Online travel reviews are emerging as a powerful source of information affecting tourists' pre-purchase evaluation of a hotel organization. This trend has highlighted the need for a greater understanding of the impact of online reviews on consumer attitudes and behaviors. In view of this need, we investigate the influence of online hotel reviews on consumers' attributions of service quality and firms' ability to control service delivery. An experimental design was used to examine the effects of four independent variables: framing; valence; ratings; and target. The results suggest that in reviews evaluating a hotel, remarks related to core services are more likely to induce positive service quality attributions. Recent reviews affect customers' attributions of controllability for service delivery, with negative reviews exerting an unfavorable influence on consumers' perceptions. The findings highlight the importance of managing the core service and the need for managers to act promptly in addressing customer service problems.

An efficient online secondary path estimation for feedback active noise control systems

In practical cases for active noise control (ANC), the secondary path has usually a time varying behavior. For these cases, an online secondary path modeling method that uses a white noise as a training signal is required to ensure convergence of the system. The modeling accuracy and the convergence rate are increased when a white noise with a larger variance is used. However, the larger variance increases the residual noise, which decreases performance of the system and additionally causes instability problem to feedback structures. A sudden change in the secondary path leads to divergence of the online secondary path modeling filter. To overcome these problems, this paper proposes a new approach for online secondary path modeling in feedback ANC systems. The proposed algorithm uses the advantages of white noise with larger variance to model the secondary path, but the injection is stopped at the optimum point to increase performance of the algorithm and to prevent the instability effect of the white noise. In this approach, instead of continuous injection of the white noise, a sudden change in secondary path during the operation makes the algorithm to reactivate injection of the white noise to correct the secondary path estimation. In addition, the proposed method models the secondary path without the need of using off-line estimation of the secondary path. Considering the above features increases the convergence rate and modeling accuracy, which results in a high system performance. Computer simulation results shown in this paper indicate effectiveness of the proposed method.