In adaptive control the goal is to deal with systems that have unknown and/or time-varying parameters. Adaptive control techniques have been developed since 1950’s and most results were proven in the cases when the time-variations were non-existent or slow. However the results pertaining to systems with fast time-variations are still limited, in particular, when it comes to plants with unstable zero dynamics. In this work we adopt the controller design technique from the area of gain scheduling, where the time-varying parameter is assumed to be measurable. We propose the design of a nonlinear periodic controller, where in each period the state and parameter values are estimated and an appropriate stabilizing control signal is applied. It is shown that the closed loop system is stable under fast parameter variations with persistent jumps: the trajectory of the closed loop state in response to the initial condition is bounded by a decaying exponential plus a gain times the size of the noise. Our approach imposes several constraints on the plant; however, we show that there exists at least one interesting class of systems, which includes plants with unstable zero dynamics, that can be stabilized by our controller.

This thesis is concerned with the analysis of the control design to the nonlinear networked control systems (NCSs). Ignoring the network connection and cascading actuators, the plant and sensors together, a sampled-data system is obtained. The stabilization problem of nonlinear sampled-data systems is considered under the low measurement rate constraint. Dual-rate control schemes based on the emulation design and discrete-time design approaches respectively are proposed that utilize a numerical integration model to approximately predict the current state of the plant. It is shown that using the dual-rate control schemes, input-to-state stability property will be preserved for the closed loop sampled-data system in a practical sense. On the other hand, the networked realization of nonlinear control systems is studied and a model-based control scheme is addressed as a solution to reduce the network traffic and resultantly, to attain a higher performance. The NCSs are modeled as continuous-time systems and sampled-data systems, respectively. Under the proposed control scheme, a tradeoff between satisfactory control performance and reduction of network traffic can be achieved. It is shown that by using the estimated values, generated by the plant model, instead of true values of the plant, a significant saving in the required bandwidth is achieved and this makes possible stabilization of the plant even under slow network conditions.

No description available.

International Telemetering Conference Proceedings / October 28-31, 1996 / Town and Country Hotel and Convention Center, San Diego, California / The traditional use of active RC-type filters as anti-aliasing filters in Pulse Code Modulation (PCM) systems is being replaced by the use of Digital Signal Processing (DSP) filters, especially when performance requirements are tight and when operation over a wide environmental temperature range is required. In order to keep systems more flexible, it is often desired to let the DSP filters run asynchronous to the PCM sample clock. This results in the PCM output signal being a sampling of the output of the DSP, which is itself a sampling of the input signal. In the analysis of the PCM data, the signal will have a periodic repeat of a previous sample, or a missing sample, depending on the relative sampling rates of the DSP and the PCM. This paper analyzes what effects can be expected in the analysis of the PCM data when these anomalies are present. Results are presented which allow the telemetry engineer to make an effective value judgment based on the type of filtering technology to be employed and on the desired system performance.

Les méthodologies de l'automatique ont joué au cours des dernières décennies un ´r^ole essentiel au sein de nombreux secteurs technologiques avancées. Cependant, de nombreuse questions restent ouvertes. Parmi celles-ci, celles concernant la stabilité et la stabilisation de systèmes non linéaires sont d'intérêt primordial. Afin de stabilizer un système (physique ou non), il est nécessaire de capter et interpreter en temps réel les informations hétérogènes caractérisant son fonctionnement afin intervenir efficacement. Actuellement ces informations ne sont pas captées en temps continu, mais de façon synchrone ou asynchrone et ceci est valable aussi pour les actuateurs. De façon très naturelle, on définit donc un système hybride, caractérisé par des dynamiques à la fois discrètes et continues. Dans ce contexte, cette thèse est orientée au développement de nouvelles méthodologies pour la stabilisation de systèmes échantillonnés non linéaires en se focalisant sur la stabilisation de formes cascades qui se retrouvent dans de nombreuse situations concretes. Pour cela, on étudiera l'effet de l'échantillonnage sur les propriétés de la dynamique continue et l'on proposera des méthodologies pour la conception de lois de commande qui ne requièrent pas d'assumptions supplémentaires au cas continu.Enfin, on étudiera l'effet de l'échantillonnage sur des systèmes présentant de retards sur les entrées. On développera des lois de commande stabilisantes exploitant la structure en cascade induite par l'échantillonnage. Des exemples académiques illustreront les calcules des solutions et leur performances tout au long du manuscript. / Over the last decades the methodologies of dynamical systems and control theory have been playing an increasingly relevant role in a lot of situations of practical interest. Though, a lot of theoretical problem still remain unsolved. Among all, the ones concerning stability and stabilization are of paramount importance. In order to stabilize a physical (or not) system, it is necessary to acquire and interpret heterogeneous information on its behavior in order to correctly intervene on it. In general, those information are not available through a continuous flow but are provided in a synchronous or asynchronous way. This issue has to be unavoidably taken into account for the design of the control action. In a very natural way, all those heterogeneities define an hybrid system characterized by both continuous and discrete dynamics. This thesis is contextualized in this framework and aimed at proposing new methodologies for the stabilization of sampled-data nonlinear systems with focus toward the stabilization of cascade dynamics. In doing so, we shall propose a small number of tools for constructing sampled-data feedback laws stabilizing the origin of sampled-data nonlinear systems admitting cascade interconnection representations. To this end, we shall investigate on the effect of sampling on the properties of the continuous-time system while enhancing design procedures requiring no extra assumptions over the sampled-data equivalent model.Finally, we shall show the way sampling positively affects nonlinear retarded dynamics affected by a fixed and known time-delay over the input signal by enforcing on the implicit cascade representation the sampling process induces onto the retarded system. Academic examples will illustrate the computational aspects together with their performances throughout the whole manuscript.

This dissertation focuses on the study of decentralized coordination algorithms of multiple autonomous vehicles. Here, the term decentralized coordination is used to refer to the behavior that a group of vehicles reaches the desired group behavior via local interaction. Research is conducted towards designing and analyzing distributed coordination algorithms to achieve desired group behavior in the presence of none, one, and multiple group reference states. Decentralized coordination in the absence of any group reference state is a very active research topic in the systems and controls society. We first focus on studying decentralized coordination problems for both single-integrator kinematics and double-integrator dynamics in a sampled-data setting because real systems are more appropriate to be modeled in a sampled-data setting rather than a continuous setting. Two sampled-data consensus algorithms are proposed and the conditions to guarantee consensus are presented for both fixed and switching network topologies. Because a number of coordination algorithms can be employed to guarantee coordination, it is important to study the optimal coordination problems. We further study the optimal consensus problems in both continuous-time and discrete-time settings via an linear-quadratic regulator (LQR)-based approach. Noting that fractional-order dynamics can better represent the dynamics of certain systems, especially when the systems evolve under complicated environment, the existing integer-order coordination algorithms are extended to the fractional-order case. Decentralized coordination in the presence of one group reference state is also called coordinated tracking, including both consensus tracking and swarm tracking. Consensus tracking refers to the behavior that the followers track the group reference state. Swarm tracking refers to the behavior that the followers move cohesively with the external leader while avoiding inter-vehicle collisions. In this part, consensus tracking is studied in both discrete-time setting and continuous-time settings while swarm tracking is studied in a continuous-time setting. Decentralized coordination in the presence of multiple group reference states is also called containment control, where the followers will converge to the convex hull, i.e., the minimal geometric space, formed by the group references states via local interaction. In this part, the containment control problem is studied for both single-integrator kinematics and double-integrator dynamics. In addition, experimental results are provided to validate some theoretical results.

No description available.

This thesis focuses on issues related to implementing theoretical Discrete-Event Systems (DES) supervisors, and the concurrency and timing delay issues involved. Sampled-data (SD) supervisory control deals with timed DES (TDES) systems where the supervisors will be implemented as SD controllers. An SD controller is driven by a periodic clock and sees the system as a series of inputs and outputs. On each clock edge (tick event), it samples its inputs, changes states, and updates its outputs. In this thesis, we identify a set of existing TDES properties that will be useful to our work, but not sufficient. We extend the TDES controllability definition to a new definition, SD controllability, which captures several new properties that will be useful in dealing with concurrency issues, as well as make it easier to translate a TDES supervisor into an SD controller. We then establish a formal representation of an SD controller as a Moore Finite State Machine (FSM), and describe how to translate a TDES supervisor to a FSM, as well as necessary properties to be able to do so. We discuss how to construct a single centralized controller, as well as a set of modular controllers and show that they will produce equivalent output. Next, we capture the enablement and forcing action of a translated controller in the form of a TDES supervisory control map, and show that the closed-loop behavior of this map and the plant is the same as that of the plant and the original TDES supervisor. We also show that our method is robust with respect to nonblocking and certain variations in the actual behavior of our physical system. We also introduce a set of predicate-based algorithms to verify the SD controllability property, as well as certain other conditions that we require. We have created a software tool for verifying these conditions and provide the source code in the appendix. We have implemented these algorithms using binary decision diagrams (BDD). For illustrative purpose, we have produced a set of examples which fail the key conditions discussed in this thesis, as well as a successful application example based on a Flexible Manufacturing System. We also presented the corresponding FSM, translated from the example's supervisors. / Thesis / Master of Applied Science (MASc)

Reinforcement Learning has received a lot of attention over the years for systems ranging from static game playing to dynamic system control. Using Reinforcement Learning for control of dynamical systems provides the benefit of learning a control policy without needing a model of the dynamics. This opens the possibility of controlling systems for which the dynamics are unknown, but Reinforcement Learning methods like Q-learning do not explicitly account for time. In dynamical systems, time-dependent characteristics can have a significant effect on the control of the system, so it is necessary to account for system time dynamics while not having to rely on a predetermined model for the system. In this dissertation, algorithms are investigated for expanding the Q-learning algorithm to account for the learning of sampling rates and dynamics approximations. For determining a proper sampling rate, it is desired to find the largest sample time that still allows the learning agent to control the system to goal achievement. An algorithm called Sampled-Data Q-learning is introduced for determining both this sample time and the control policy associated with that sampling rate. Results show that the algorithm is capable of achieving a desired sampling rate that allows for system control while not sampling “as fast as possible”. Determining an approximation of an agent’s dynamics can be beneficial for the control of hierarchical multiagent systems by allowing a high-level supervisor to use the dynamics approximations for task allocation decisions. To this end, algorithms are investigated for learning first- and second-order dynamics approximations. These algorithms are respectively called First-Order Dynamics Learning and Second-Order Dynamics Learning. The dynamics learning algorithms are evaluated on several examples that show their capability to learn accurate approximations of state dynamics. All of these algorithms are then evaluated on hierarchical multiagent systems for determining task allocation. The results show that the algorithms successfully determine appropriated sample times and accurate dynamics approximations for the agents investigated.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第18993号 / 工博第4035号 / 新制||工||1621(附属図書館) / 31944 / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 萩原 朋道, 教授 松尾 哲司, 准教授 古谷 栄光 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM