Keywords: Learning

Keywords: Automotive applications, Transportation systems, Control applications
Abstract: Connected and automated vehicles (CAVs) have been recognized as providing unprecedented opportunities for substantial fuel economy improvement through CAV-based vehicle speed trajectory optimization (eco-driving). At the same time, the implications of the CAV operation on thermal responses, including those of engine and exhaust aftertreatment system, have not been fully investigated. To this end, firstly, a sequential optimization framework for vehicle speed trajectory planning and powertrain control in hybrid electric CAVs is proposed in this paper. Next, the impact of eco-driving and power split optimization on the engine and catalytic converter thermal responses, as well as on the tailpipe emissions is characterized. Despite an average 16% improvement in fuel economy through sequential optimization, this study shows that eco-driving slows down the thermal responses, which could unfavorably affect the tailpipe emissions.

Keywords: Automotive applications, Control Technology, Predictive control
Abstract: In this paper, an optimal and predictive multilevel control solution is developed to utilize data and capabilities provided by the Connected and Automated Vehicle (CAV) technologies to reduce fuel consumption for line-haul class 8 trucks. The longer horizon road grade and speed limit data combined with shorter horizon and time-varying traffic data on real-world driving scenarios is utilized in the developed integrated optimal control algorithms to further reduce fuel consumption with respect to a production baseline control with state-of-art Advanced Driver-Assistance Systems (ADAS) control features. Furthermore, integration of the developed control algorithms for the front vehicle automation in platoon with a rear truck is also studied. The results indicate that with the application of the developed control solutions for powertrain efficiency in both single truck automation or the automation of the front truck in platoon with a rear truck, the fuel consumption is significantly reduced over real-world driving scenarios with road grade, speed limit and traffic data. The proposed solution is implementable in real-time with the utilization of CAV technologies with the proposed multilevel control architecture.

Keywords: Optimization, Automotive applications, Simulation
Abstract: This paper is an extension of our previous study on optimizing energy-efficient speed profiles of electric vehicles. This paper investigates the influence of motor efficiency on performance, including the energy consumption and control modes through parametric studies, for a light weight military ground vehicle. A comprehensive comparison is conducted between the global optimal solutions obtained from DP with nonlinear motor efficiency, and the suboptimal solutions derived from various constant efficiencies. The global optimal solutions lead to about 2-3% lower energy consumption than the suboptimal solutions for the considered driving scenarios. It is observed that the energy consumption and control modes of the suboptimal solutions are hardly affected by the motor efficiency. For a flat road, the control modes of the global optimal solutions, characterized as Pulse and Glide, are considerably different from those of the suboptimal solutions. However, no significant difference is observed in control modes for a hilly terrain.

Keywords: Automotive applications, Sliding mode control, Fault-tolerant systems
Abstract: Cyclic torque imbalance is one of the major problems of gasoline engines that is caused due to variations between combustion cycles. Output torque generated by the engine cylinder is required to be same for each combustion cycle but variations occur in it due to disturbances in torque generation. Although, cyclic torque imbalance was sporadically studied but it was not considered a concern until measurement of consecutive engine cycles was made possible.

In this paper, an Observer Based Robust Control (OBRC) methodology is proposed to develop a novel model based unified framework for detection and mitigation of cyclic torque imbalance. Based on estimated net piston force, second order sliding mode control technique is applied to mitigate imbalance in the cyclic torque; which is generated by inducing fault in the fuel injection subsystem. The unified framework is developed by using First Principle-based Engine Model (FPEM). The proposed Observer Based Robust Control methodology is validated by MATLAB/Simulink simulations.

Keywords: Optimization, Control applications
Abstract: Electric vehicles (EVs) powered by electric motors and batteries have solved the issues related to the environmental pollution. The major concern with electric vehicles is their short range, which is due to the limited capacity of a battery installed in the system. Limited range issue can be solved or minimized through eco-driving; an optimization technique in which a driver is advised to follow a velocity profile that consumes minimum energy. In the recent past, eco-driving problem with an online requirement using the static model of a battery has been discussed. Eco-driving problem has been formulated as an optimal control problem, where the goal is to minimize vehicle power consumption only. In this work, the goal is extended to increase the life cycle of battery along with power consumption. For this purpose, battery dynamics are included along with vehicle dynamics. To increase the battery's life cycle constraints are applied to avoid the battery from getting fully charged and discharged.Problem is solved using Pontryagin's Minimum Principle (PMP), where initial and final position of a vehicle, initial velocity of a vehicle and total trip time is fixed. Optimal torque, optimized velocity trajectory and constrained state of charge during a trip are generated that in turn minimize the energy consumption of the battery. The results have shown that constrained case is approximately 7 percent more efficient in terms of energy consumption than an unconstrained case. Reduction in energy consumption, coupled with battery constraints result in proper charging and discharging of the battery thereby extending battery life.

Keywords: Transportation systems, Autonomous systems, Sensors
Abstract: This paper proposes a vision-based perception system of using a monocular camera to detect the high-level features in the form of road markers and driving lanes for providing reliable visual measurements. For road marker recognition, the template matching-based method is applied to recognize the marker type and estimate the corresponding position and orientation as well. For driving lane detection, the proposed gradient orientation consistency combined the Inverse Perspective Mapping (IPM) spatial constraints is used to initialize the clearly visible lanes in the initial detection. Finally, the particle filter is employed to integrate the visual measurements, inertial measurement unit (IMU), and GPS complementarily for correcting GPS errors. The proposed vehicle localization system is evaluated in the real driving scenario of the standard campus environment within varying illumination condition.

Keywords: Aerial robotics
Abstract: A novel omnidirectional aerial robot is introduced in this paper. It is driven by four 2-DOF tilting modules mounted at four vertices of a regular tetrahedron. This configuration ensures that at least three rotors remain on one horizontal plane while the frame rotates at any angle. The robot is thus capable of flying at any orientation in SE(3). This significantly expands application areas of the robot. Reliability is enhanced thanks the over-actuation design and configuration of the rotors. Dynamic equations of motion are derived based on the mechanical design. A control strategy is represented that allows decoupling of rotational and translational dynamics. A conceptual method for control allocation is also proposed. This takes into consideration the input saturation as well as other physical constraints. Furthermore, robustness is provided by re-configuration of rotors if some actuator encounters failures. Preliminary simulation and experiments demonstrate feasibility of the novel design and capability of this aerial robot.

Keywords: Aerospace applications, PID control, Modeling
Abstract: This paper proposes a design method for a two-degrees-of-freedom tracking controller for a quadrotor to ensure trajectory following. The controller consists of two parts. A flatness-based feedforward controller ensures trajectory tracking of the undisturbed vehicle and a feedback controller composed of an altitude controller and a position and attitude controller in a cascaded structure compensates the tracking error due to disturbances and model uncertainties. The feedback controller is designed with the root locus method based on the linearized model of the quadrotor. Experiments are made in a testbed under windy conditions with a bounded windforce. The results show that the quadrotor with the proposed controller follows a given trajectory closely with small, bounded deviations, which can be estimated in advance with the model of the quadrotor.

Keywords: Navigation, Aerial robotics, Kalman filtering
Abstract: In indoor applications of unmanned multirotors, absolute localization with high availability is necessary to enable autonomous flights. Due to the fast dynamics of multirotors, (smooth) estimates of position, velocity and orientation are required in real-time and at high update rates. This paper addresses the indoor state estimation problem by means of a loosely-coupled integration of an inertial navigation scheme with position measurements of a localization system. State propagation is performed using a Cartesian implementation of the strapdown integration of inertial measurements. An error-state extended Kalman filter is then applied, compensating the delay in availability of position measurements. Equations are derived in detail, and parameters for initialization and parametrization are given. For performance discussion, experiments are performed and discussed using different filter parameters, variation of the assumed measurement delay, and artificial outages in position measurements. Results demonstrate the suitability of the proposed method, showing mean errors in position estimates below 5cm, and realizing an error reduction of up to 50% by consideration of delays.

Keywords: Autonomous systems, Kalman filtering, Aerospace applications
Abstract: A Kalman filter-based wind estimate method is presented in this paper for wildfire monitoring. Wind behavior is one of the key factors in fire propagation, especially on the early stage of fire. This quadrotor UAV-based method is proposed to estimate the lateral wind in the fire environment only with a standard autopilot sensor suite on board. A Kalman filter is designed to estimate the wind vector based on the vehicle's hovering inclined angle in the wind environment. The designed Kalman filter can reduce the influence of sensor noises and improve the accuracy of the wind estimation. The effectiveness of the algorithm is demonstrated in a simulation environment.

Keywords: Linear parameter-varying systems, Aerospace applications, Sensor fusion
Abstract: This paper studies the placement of sensors to achieve optimal vibration suppression for a flexible BWB (Blended-Wing-Body) airplane wing. For a given flight speed range, vibration behaviors of the wing structure are evaluated by guaranteed H₂ performance of the closed-loop system with the H₂ LPV controller. Candidate sensor locations are identified on each wing, and the optimal sensor placements are found by globally searching through all possible combination over those candidate locations. With the LPV modeling of a flexible wing and H₂ controller synthesis conditions, simulation results provide the optimal sensor locations for a given number of sensors. In addition the trade-off between optimal system performance and number of sensors can also be attained.

Keywords: Aerospace applications, Autonomous systems, Optimization
Abstract: The guidance design of aerobatic aircraft in air-race is to ensure several stringent terminal conditions at the race gates as well as to satisfy bounds on control and structural load factor. Also, it should be robust in presence of uncertainties in aerodynamic parameters. All these lead to it becomes a challenging problem. To meet these requirements, the generalized model predictive static programming (G-MPSP) based suboptimal guidance design with hard terminal and input inequality constraints is presented in this paper. Penalty function approach is followed with the G-MPSP algorithm to ensure input constraints. To demonstrate the robustness of the proposed guidance algorithm, the simulation results are carried out considering aerodynamic parameter uncertainties.

Keywords: Control applications, Filtering, Linear robust control
Abstract: I will talk about my own experience in materializing a sound processing patent to a factory-made LSI chips. There are various unexpected steps toward real applications, and useful lessons to be shared with the audience.

Keywords: Complex systems

Keywords: Control Technology, Smart structures, Cyberphysical systems
Abstract: Society 5.0 is the next-generation society,proposed in the 5th Science and Technology Basic Plan by the Cabinet Office in Japan. Society 5.0 is technologically defined as a system of systems (SoS), where many systems (e.g. energy management systems, road transport systems, etc.) are connected through the Internet to achieve some global requirements (e.g. reducing carbon emissions). In this article, we will review the concept of Society 5.0 and key technologies for Society 5.0. We then introduce a research project of Society 5.0 in Kitakyushu, Japan. The project team includes not only engineering researchers but also an economist, a social scientist, engineers from industry, and people from government. In this article, I will point out difficulties, which we may always face in a big project, based on my experiences in our collaborative research project.

Keywords: Biosystems, Biotechnology

Keywords: Linear robust control, Control architectures, Health and medicine
Abstract: This paper will be given via a video presentation.

Keywords: Verification and validation, Linear systems, Simulation
Abstract: A simulation-based method to analyze the performance of a class of uncertain nonlinear systems controlled by linear state-feedback controllers is proposed in this work. The aim is to certificate numerically the ability of a given linear controller to steer a nonlinear system from its initial state set to its desired target set without violating the state and input constraints. To achieve that, first, we propose a new reachability analysis method for discrete-time systems. Then we establish a set-membership check-list to evaluate the expected performance of the linear controller. A case-study borrowed from the literature is presented to illustrate the principle of the proposed approach and to show its effectiveness.

Keywords: Control applications
Abstract: Normally, it is observed that when the system is perturbed, there is need to design an advanced controller such as H infinity, Quantitative feedback technique (QFT) , Sliding mode control etc. However, in this short paper, it is shown that for some class of systems even though the model of the system is perturbed, the fixed PD control is possible to design without using advanced control techniques. The beauty of the proposed approach is that the proposed approach gives direct formulae for the controller design. The proposed approach is validated on a Qube Servo system which shows satisfactory results in simulation and also on a hardware setup.

Keywords: Linear systems, PID control, Adaptive control
Abstract: Smart strong stability system is one of the safe control systems, whose open-loop and closed-loop gains are designed to be 1. In the steady state, the smart strong stability system becomes a quasi-open-loop system. In other words, the feedback signal becomes almost zero in the steady state. Therefore, even if the accident that the feedback signal becomes zero occurs, the smart strong stability system can keep the safety. Although the proposed system was derived by generalized minimum variance control (GMVC) in the previous researches, this paper challenges the derivation of the proposed system through self-tuning PID control.

Keywords: Linear robust control, Optimization, Identification
Abstract: Many car manufacturers are equipping their vehicles with advanced systems to win the race towards autonomous vehicles. In order to ensure multiple objectives, different embedded systems influencing differently the same physical variable are often implemented in the same vehicle. Most of car manufacturers tend to activate one system at a time, especially when the low-level control of actuators is developed by different equipment suppliers, making the inner dynamics uncertain. However, this limits the potential of the vehicle. This paper discusses the feasibility of coordinating these systems when activated at the same time, while ensuring an acceptable robustness regarding the uncertain dynamics. Results showed good performance in severe maneuvers when combining robust control synthesis and optimization-based control allocation algorithms. Therefore, several embedded systems may be activated at the same time to expand the overall potential of the vehicle and deal with more difficult situations.

Keywords: Linear systems, Linear robust control, Optimization
Abstract: This article considers the problem of sub optimal regulator design for linear systems subject to additive but bounded disturbances. The information regarding the bounds is used to formulate another optimal control problem. The system used in the latter problem is the original system subject to worst case of disturbance. Then the latter optimal control problem is solved using Krotov sufficient conditions via selection of the so-called Krotov function. The selection of Krotov function is made such that the iterative computation of solutions (which is required while employing these conditions) is avoided. The resulting controller, which is sub-optimal for the original problem, is comprised of a state-feedback term and a time-varying term which compensates for the effect of disturbances. The numerical examples demonstrate the application of developed methodology.

Keywords: Linear systems, Observers
Abstract: This paper considers the stabilization problem of singular systems represented by an equivalent system involving input derivatives. Stabilization is achieved by eigenvalue assignment after a subsequent elimination process of input derivatives. First, the connection between the class of systems with derivative inputs and singular systems is analyzed. Then an elimination procedure for input derivatives is given, which transfers the derivative terms from the state equation to the output equation allowing stabilization by state feedback to be performed under a modified controllability condition. The design of observer for the transformed system with input derivatives appearing in the output equation is also considered. It is shown that a similar elimination process can be performed directly in the design procedure of the observer. An immediate benefit of the new approach is the fact that it can be employed to systems that naturally admit input derivatives in their state space models. This is further elaborated in the paper and illustrated as part of design examples.

Keywords: Discrete event systems, Verification and validation, Transportation systems
Abstract: The development of supervisory controllers for cyber-physical systems is a difficult and error-prone process. Supervisor synthesis enables control designers to automatically synthesize a correct-by-construction supervisor from a model of the plant in combination with a model of the control requirements. To validate the behavior of the synthesized supervisor, model simulation is commonly used. While model simulation is a powerful tool for validation, it offers only a partial analysis. Aspects related to the execution of the supervisor on the hardware and communication with subsystems, such as human machine interfaces, cannot be validated with model simulation. To bridge the gap between model simulation and realization, hardware-in-the-loop (HIL) simulation can be performed after model simulation and before implementation. For HIL simulation, the controller realization and its subsystems are connected to a model of the plant.

The purpose of this paper is to propose an engineering method that combines synthesis-based engineering with HIL simulation. Models created for synthesis are refined and re-used to obtain models for model simulation and HIL simulation. To illustrate this method, a case study is presented that demonstrates its application to the Prinses Marijke complex, consisting of two waterway locks and a storm surge barrier. The necessary models have been developed and implemented in a HIL set-up.

Keywords: Simulation, Discrete event systems, Optimization
Abstract: In this paper, we consider the ranking and selection problem of selecting the optimal subset from a finite set of alternative designs. Given the total simulation budget constraint, we aim to maximize the probability of correctly selecting the top-m designs. In order to further improve the selection efficiency, we incorporate the information from across the domain into quadratic regression equation. Under some common assumptions in most regression based approaches, we propose an approximately optimal rule that determines the design locations need to be simulated and the number of simulation replications allocated to the selected designs. Numerical experiments demonstrate that our approach dramatically improves the selection efficiency on some typical selection examples compared to the existing approaches.

Keywords: Discrete event systems, Optimization, Renewable Energy
Abstract: In this paper, we propose a new optimization method to simultaneously maximize the average return and to minimize the reward variance of a stochastic dynamic system. This problem cannot be formulated as a standard Markov decision process (MDP) since the optimization criterion is a combined metric with mean and variance. Traditional methods, such as dynamic programming, are not valid. We resort to the sensitivity-based optimization theory and propose a new method to solve this problem. We derive a performance difference formula which quantifies the difference of the mean-variance combined metrics under any two different policies. Some optimality structures of this problem are also derived, which can be utilized to further develop iterative algorithms to optimize the mean-variance metrics.

Keywords: Complex systems, Complex networks, LMIs
Abstract: This paper investigates the event-triggered H-infty control problem for network-based uncertain stochastic linear systems with exogenous disturbance. In order to reduce the amount of signal transmission and update, the exogenous disturbance parameter is introduced into the sampled-data based event-triggered scheme (ETS). Under the state feedback control, a time-delay closed-loop uncertain stochastic system model is further developed. Then by using Lyapunov-Krasovskii functional method, a set of sufficient conditions in terms of linear matrix inequalities (LMIs) are given to guarantee the H-infty performance of the resulting time-delay closed-loop system. Subsequently, we present conditions to design the parameters of the state feedback gain and the ETS. Finally, an example is provided to show the effectiveness of the proposed method.

Keywords: Discrete event systems, Switched systems, Actuators
Abstract: This paper is considered with semi-global stabilization of discrete-time linear systems subject to actuator saturation based on event-triggered control (ETC) and self-triggered control (STC). First, a novel ETC algorithm is proposed by utilizing properties of the discrete-time parametric Lyapunov equation. Moreover, the corresponding STC, in which the next control law depends on the previous triggered states, is designed to avoid continuous monitoring system states. The designed control algorithms not only save energy but also provide a very clear relationship between the only design parameter and the minimal inter-event time, which can be easily used to change regularly inter-event times (IET) by adjusting the only design parameter and allows us to find easily a tradeoff between IET and the control performance. Finally, the designed ETC and STC are used to stabilize the inertia wheel pendulum system. Numerical simulations show the effectiveness of the proposed algorithms.

Keywords: Hybrid systems, Mechatronic systems, Mechanical systems
Abstract: Abstract—Integrated atomic force microscope (AFM) and confocal laser scanning microscope (CLSM) can quickly obtain the three-dimensional (3-D) surface of the sample in large scanning range and recover the region of interesting (ROI) in nanoscale resolution. However, the traditional cooperative algorithm for integrated microscopes occupies too much scanning time. In this work, we develop a novel cooperative algorithm for the integrated microscopes to reduce scanning time of AFM and achieve higher scanning speed. First, the calibration of the microscopes will be implemented. Next, CLSM starts a large range scan first and then define the region of interesting (ROI) by edge detection. And then, the scan regions of the AFM are arranged based on the ROI and adaptive scanning region method is proposed to reduce the scanning time. Furthermore, variable speed scanning based on the height information obtained from CLSM image is applied to increase the AFM scanning speed. Finally, the scanning images obtained from AFM and CLSM are merged together. A series of experimental results show that proposed cooperative algorithm can save approximately 69.2% of scanning time compared with that obtained by traditional cooperative algorithm.

Keywords: Autonomous systems, Automotive applications, Verification and validation
Abstract: As the decision making responsibilities of autonomous vehicles increase, they will be expected to navigate complex, unstructured environments such as traffic circles. These environments necessitate effective safety control algorithms. Control barrier functions provide such a tool for guaranteeing system safety by ensuring that control actions render a given safe set forward invariant. However, finding an appropriate control barrier function is challenging. To alleviate this challenge, we consider a nominal evasive maneuver for the system. Then a control barrier function is designed by considering the closed loop dynamics resulting from this hypothetical evasive maneuver. Using this approach, in this paper, we propose a control algorithm to navigate an autonomous vehicle through a traffic circle in the presence of other vehicles. The synthesized control barrier function is able to simultaneously ensure lane keeping while avoiding collisions with other vehicles. The solution approach is then physically demonstrated on the Robotarium remote access testbed.

Keywords: Autonomous systems, Control applications, Linear systems
Abstract: The implementation of connected and automated vehicle (CAV) technologies enables a novel computational framework for real-time control actions aimed at optimizing energy consumption and associated benefits. Several research efforts reported in the literature to date have proposed decentralized control algorithms to coordinate CAVs in various traffic scenarios, e.g., highway on-ramps, intersections, and roundabouts. However, the impact of optimally coordinating CAVs on the performance of a transportation network has not been thoroughly analyzed yet. In this paper, we apply a decentralized optimal control framework in a transportation network and compare its performance to a baseline scenario consisting of human-driven vehicles. We show that introducing of CAVs yields radically improved roadway capacity and network performance.

Keywords: Optimization, Transportation systems, Autonomous systems
Abstract: In this paper, we provide a decentralized optimal control framework for coordinating connected and automated vehicles (CAVs) in two interconnected intersections. We formulate a control problem and provide a solution that can be implemented in real time. The solution yields the optimal acceleration/deceleration of each CAV under the safety constraint at "conflict zones," where there is a chance of potential collision. Our objective is to minimize travel time for each CAV. If no such solution exists, then each CAV solves an energy-optimal control problem. We evaluate the effectiveness of the efficiency of the proposed framework through simulation.

Keywords: Autonomous systems, Planning, Control Technology
Abstract: On-demand air mobility has gained attention in the recent years as a promising future transportation mode for daily commutes. In this paper, we propose a hierarchical motion planning system for on-demand passenger-carrying air vehicles. Specifically, we consider a path planning problem with requirements for flight efficiency, system safety, and passenger comfort. Prior to the execution of the mission, we simplify a given city map into a grid world and estimate the transfer time among the cells in the presence of no-fly zones and air traffic congestion. Then, we consider system reachability and apply A* search to find a time-optimal path in the map. To ensure that the system follows a feasible trajectory consistent with the off-line plan after take-off, we use model predictive control to track the waypoints on the path and minimize the energy consumption during flight. In order to account for collisions unforeseen in off-line planning, we further adopt a collision avoidance module in the system, which optimizes a new waypoint near the predicted collision point and generates an avoidance trajectory through model predictive control. At last, we demonstrate the effectiveness of the hierarchical planning system under different flight scenarios in a simulated urban environment.

Keywords: Autonomous systems, Transportation systems, Mobile Robots
Abstract: Investigations about backward motion of autonomous Truck-Trailer is interesting because the backward motion system is harder than forward motion system. The system is hard to be controlled and the possibility of jack-knife effect to occur is high. This paper presents line path following control of an autonomous truck-trailer backward motion. Control architecture and design control is proposed. Design control of line path following control is divided into three steps, namely truck-trailer orientation, steering angle control, and position control of steering angle. The input of the system is traction velocity and steering angle as determinant of the truck-trailer system steering direction. Using truck-trailer oriented algorithm, backward motion control is focused on its head so that the truck-trailer motion will follow the desired path. Line path following control is designed using Lyapunov Stability approach to find the system stability. The results of the simulation studies showed that the system has reached asymptotic global stable condition on the origin of error, which means that the distance and orientation of the head truck to the path is zero so the motion of the system is on the path.

Keywords: Health and medicine
Abstract: Because of continuous technical progress over the last decades artificial pancreas (AP) systems have recently become reality for patients with type 1 diabetes (T1D) with first AP devices entering the market. However, all AP systems commercially under development are so-called hybrid closed loop systems, meaning that only the basal insulin is adjusted by the algorithm, whereas the meal boluses still have to be triggered manually by the patient. Fully closed loop systems for T1D that also administer meal boluses autonomously are so far mainly of academic interest. The biggest drawback of fully closed loop systems is that they can only give an insulin bolus as soon as a meal is detected from the recorded data, leading to significant delays in insulin delivery (and therefore more pronounced peaks in postprandial glucose) and/or an increased risk of hypoglycemia (in case of false positive detections).

Contrary to almost all publications on the topic, the current paper investigates the feasibility of fully closed loop systems for patients with type 2 diabetes (T2D). Since glycemic variability tends to be significantly lower in T2D and since glucose dynamics are somewhat slower, fully closed loop systems seem in fact more favorable for this patient group than for T1D. The paper investigates the performance of available meal detection algorithms from the literature (developed for T1D patients) and modifies them for application in T2D. Additionally, the paper proposes a simple fully closed loop algorithm for T2D patients including a module for automatic meal bolusing and demonstrates its feasibility and safety.

Keywords: Biotechnology, Control applications, Predictive control
Abstract: Type 2 diabetes (T2D) accounts for 90% of the diabetes patients, and more than 60% of patients on insulin fail to reach treatment targets. The main reasons include low adherence to treatment, caused by, among other reasons, complexity and fear of overdosing.

In insulin titration, pre-breakfast SMBG measurements are used for dose calculations. However, glucose values may be lower between the daily SMBG measurements. Not considering this may lead to over-titration. We propose a dose guidance algorithm to identify individualized optimal dosing of long acting insulin for T2D patients.

We use a compartment model of fasting glucose and long acting insulin dynamics in T2D to develop a model predictive control (MPC) based decision support system. To promote safety, the controller optimizes the daily dose based on predicted fasting glucose values with faster sampling than available data allows. We test the performance of the controller with respect to safety and efficacy in a simplified in silico environment and compare the results with standard of care dose guidance. The aim of this paper is to illustrate the importance of fast sampling in prediction between the slow input samples. This importance is observable in the results, which are indicative of the potential of such an approach.

Keywords: Estimation, Filtering, Kalman filtering
Abstract: In this paper, we consider nonlinear state estimation in the U-loop reactor for single-cell protein (SCP) production. The model of the U-loop reactor is a mixture of stochastic partial differential equations and stochastic differential equations which are stiff. By a typical finite-volume spatial discretization, the resulting system of stochastic differential equations for numerical simulation and state estimation has 83 states. We investigate and discuss the continuous-discrete EKF for state estimation in this high-dimensional and stiff continuous-discrete-time system.

Keywords: Health and medicine, Iterative learning control, Robotics
Abstract: Effective training requires high muscle forces potentially leading to training-induced injuries. Thus, continuous monitoring and controlling of the loadings applied to the musculoskeletal system along the motion trajectory is required. In this paper, a norm-optimal iterative learning control algorithm for the robot-assisted training is developed. The algorithm aims at minimizing the external knee joint moment, which is commonly used to quantify the loading of the medial compartment. To estimate the external knee joint moment, a musculoskeletal lower extremity model is implemented in OpenSim and coupled with a model of an industrial robot and a force plate mounted at its end-effector. The algorithm is tested in simulation for patients with varus, normal and valgus alignment of the knee. The results show that the algorithm is able to minimize the external knee joint moment in all three cases and converges after less than seven iterations.

Keywords: Predictive control, Biotechnology
Abstract: In this paper, we consider economic optimal control of single-cell protein (SCP) production in a U-loop reactor. The model of the U-loop reactor contains both ordinary and partial differential equations. Consequently, the optimal control problems are large-scale. The optimal operating profile for the SCP production is an unstable attractor. Therefore, we consider two simultaneous collocation-based approaches for solving the optimal control problems. We implement these two approaches in C, and we use IPOPT to solve the involved nonlinear program (NLP). Finally, we present a performance study that demonstrates the feasibility of solving economic optimal control problems that involve the U-loop reactor in real-time.

Keywords: Robust control, Transportation systems, Control applications
Abstract: High speed trains are being developed rapidly world-wide, particularly in China. A train is powered by several pantograph-catenary systems. The power transmission efficiency is the central issue in this system. To this end, a high precision control of the contact force between the catenary and the pan-head is a must. However, owing to the feature of the catenary the stiffness varies periodically and the frequency is rather high for a high-speed train. This fast-varying stiffness brings about a big oscillation in the contact force and must be actively controlled. This problem has not been well resolved up to now. In this talk, a phase-shaping approach, combined with the idea of disturbance observer, is proposed and validated. It is verified that this method outperforms all existing technologies.

Keywords: Control applications, Complex networks

Keywords: Control applications, Identification
Abstract: One of the key parameters of a permanent-magnet-synchronous-motor (PMSM) servo system is the inertia of the system. Since it changes during a control task, it is required to be identified in a real-time fashion so as to achieve satisfactory control performance. This paper presents an improved identification method based on a recursive least-squares (RLS) algorithm with a varying forgetting factor (VFF-RLS). This method yields a good balance between identification precision and convergent speed.

Keywords: Control applications, Intelligent systems
Abstract: The contradiction between supply and demand of resources and energy has become increasingly prominent. To alleviate the shortage of resource supply, it is urgent to develop deep geological drilling and unconventional resources/energy. However, traditional drilling technology can no longer meet the demands of deep geological drilling. This talk presents the intelligent control system of drilling process for complex geological environment. In the intelligent control systems, a three-dimensional spatial model for the formation characteristics field is established to describe the complex geological environment. And based on the model, the drilling trajectory is designed to obtain the optimal drilling trajectory and the drilling control constraints. Then, complex coupling control loops-based intelligent control is introduced to enhance the efficiency and guarantee the safety of drilling process. For remote monitoring and abnormal state warning of drilling process, a state monitoring technology with three-dimensional virtual visualization is investigated. Simulations and experiments are developed to verify the system technologies.

Keywords: Smart grid
Abstract: This paper studies fifth-order filters to realize reduction of filter size for grid-connected inverters. Those filters are expected to achieve better performance with smaller filter size, since they have steeper characteristics than conventional and lower-order ones. The parameters design criteria of those filters are also introduced. The comparative analysis regarding the conventional LCL filter and the proposed fifth-order filters have been presented.

Keywords: Control applications, Nonlinear systems, Distributed parameter systems
Abstract: This paper presents a novel active disturbance suppression approach based on equivalent-input-disturbance (EID) method for single link flexible joint manipulator (FJM) with dead zone and unmodeled dynamics. Furthermore, it carries out the trajectory tracking control of this system without using velocity measurements. The stability is proved based on the separation theorem and the small gain theorem.

Keywords: Predictive control, Control applications, Optimization
Abstract: In this paper, we address an online sensor transmission scheduling of networked control systems (NCSs). The scheduler decides some sensor values to be sent to a controller via network so as to reduce communication cost while considering control performance of NCSs. The scheduling is executed in the controller side, which does not require computation load in the sensor side. The effectiveness of the proposed method is verified via a numerical example.

Keywords: Control applications, Optimization
Abstract: An ensemble clustering algorithm based on affinity propagation clustering (APC) algorithm is proposed in the paper. Different base clustering results are obtained by using APC algorithm under the selected preference parameter. However, some base clusters have little effect on the final clustering result, the “good” base clusters should be selected from all base clusters. Then the diversity and coverage ratio are applied to the judgement of the homogeneity of generated base clusters. Finally, the evidence accumulation clustering (EAC) algorithm is used for obtaining the co-association matrix from the selected base clusters, and generating final the clustering result. Experimental results on two data sets demonstrate that the clustering ensemble based on the refined co-association matrix outperforms some state-of-the-art clustering ensemble schemes.

Keywords: Autonomous systems, Nonlinear systems, Control applications
Abstract: In this paper, we consider the multi-agent formation problem with trajectory tracking and obstacle avoidance. Out of all the controllers proposed to solve the multi-task problem, null-space based controller stood out for its ability of task priority appointment, priority interchangeability, and the simplicity in Lyapunov stability analysis. However, the advantages above are mainly based on the assumption that the considered tasks are mutually non-conflicted, which is not always the case. We propose a null-space based controller that takes the lower-rank task’s control signal as the higher-rank controller’s input, and redesign higher-rank controller so that the eliminated control signal is partially compensated and the tasks’ aggregated error is decreased. We also design a dual null-space controller in order to apply virtual leader for path planning, which enables obstacle avoidance and tracking while the target tracking point is in an inadmissible condition. We verify the performance of formation and tracking control with and without dynamic obstacles, and show the decrease in aggregated error compared with the traditional null-space controller.

Keywords: Linear parameter-varying systems, Mechanical systems, Nonlinear systems
Abstract: This paper gives the formulation of a parameter-dependent input shaping filter suitable for residual vibration reduction in motion control of non-LTI systems. The method is based on the assumption that the vibratory dynamics can be represented by a second order linear ODE with time-varying coefficients that can be estimated during motion. The control implementation involves a time-varying filter function that is convolved with the system command/actuation signal in real-time. Discrete-time implementation and performance aspects are studied by numerical simulation. Numerical results for a benchmark mechanical system are presented that reveal limitations due to non-linear dynamics impacting on the accuracy of the LTV assumption. Results from implementation on a flexure-jointed X-Y motion stage mechanism are shown that confirm the applicability and effectiveness of the method in achieving reduced vibration and settling time for step-wise motion tasks.

Keywords: Sliding mode control, Nonlinear systems, Algebraic/geometric methods
Abstract: Terminal sliding modes (TSM) and their variants have been applied in diverse systems, most notably robotic manipulators. The primary motivations for their application are that they allow for fast and finite-time convergence of the states being controlled to their desired values. In this paper, the design of TSM controllers is considered when the available control magnitude is bounded. In such conditions, it will be shown, by estimating the domain of stability in the system state-space, that not all initial conditions permit the occurrence of terminal sliding mode. Using this method, design guidelines are provided for the selection of the TSM parameters that can allow larger deviations to be handled, when the control is bounded.

Keywords: Nonlinear systems, Simulation, Transportation systems
Abstract: This paper presents a simple model of the heave (up-and-down) dynamics of an air cushioned vehicle. A set of nonlinear differential equations are extended to the case of a bagged air skates using a Forchheimer porosity approximation. The model exhibits stability under atmospheric pressure, and a tendency towards instability in near-vacuum conditions. Inner estimates of regions of attractions are found to verify stability in atmosphere, and track simulation shows system disturbance rejection under a drifting subtrack height and gaps.

Keywords: Nonlinear systems, Genetic algorithms, PID control
Abstract: The aim of this investigation was to design a genetic algorithm-optimised nonlinear pseudo-derivative feedback (NPDF) controller for the regulation of a high-speed, supertall building elevator. The efficacy of the NPDF controller was compared with that of a nonlinear proportional-integral-derivative controller, as well as the conventional equivalents of both architectures. Although all controllers successfully ensured that the elevator car arrived within the required time, the NPDF controller was found to be superior in terms of solution fitness, arrival time, rise time, settling time, disturbance rejection and sensitivity to variations in cabin occupancy. Additionally, the NPDF controller was the only control scheme to meet all performance specifications for each of the various cabin masses considered in this paper.

Keywords: Modeling, Linear systems

Keywords: Linear systems, Identification

Keywords: Linear systems, Control applications

Keywords: Fault detection/accomodation
Abstract: This paper considers the problem of robustly identifying m intruders in a network consisting of n cooperative agents which are subject to unknown disturbances. First, a distributed system model is introduced so that the relationship between agents, the attacks and unknown disturbances can be captured. Next, the distributed identification scheme is formulated as a spectral assignment problem and necessary filter gains are obtained through a carefully constructed linear system of equations. Necessary and sufficient conditions to decouple the unknown disturbances from the agents residual generators are derived in terms of filter gain matrices. It is shown that the problem of discriminating between unknown disturbances and attacks in a distributed system under consensus dynamics can be reduced to the problem of determining a set of constraints on the spectrum of the residual generator coefficient matrices. The approach is illustrated through an example.

Keywords: Fault detection/accomodation, Neural networks, Machine learning
Abstract: Increasing complexity of industrial processes has made statistical methods for process monitoring and diagnosis a more attractive alternative to model-based methods. A primary reason is that statistical approaches can be formulated to rely less on process knowledge. Since multivariable processes can exhibit complex, nonlinear dynamics, there is a need for methods capable of diagnosing nonlinear process data. A Monte Carlo simulation was conducted on a numerical model of the quadruple tank process (QTP) - a novel multivariate nonlinear process. The simulation was designed so that the QTP exhibited bipartite nonlinear behavior. Reference data obtained from the simulation was used to obtain principal component analysis (PCA) and autoencoder (AE) models. The models generated residuals that were used to monitor the condition of the process. The results showed that AEs, which have nonlinear functionalities, performed better than PCA models at generating residuals.

Keywords: Fault detection/accomodation, Estimation, Sliding mode control
Abstract: This study is concerned with the design of a distributed fault estimation scheme for linear multi-agent systems subject to actuator faults. For each agent, a sliding-mode observer-based estimator module is proposed that uses the available local relative output measurements and the information transmitted from the neighboring agents. By considering the H1 performance index and using the linear matrix inequality technique, the parameters of the observers are designed such that the fault estimation is robust against disturbances. The proposed method improves the existing fault estimation techniques in terms of both complexity and performance. A simulation example is presented to illustrate the effectiveness of the proposed methodology.

Keywords: Fault detection/accomodation, Switched systems, Linear parameter-varying systems
Abstract: The problem of simultaneous state and fault estimation for switched linear parameter varying systems with dwell-time constraint is addressed in the discrete-time domain in this paper. The polytopic dependence property is assumed for the parameter variations. By constructing a suitable dwell-time-dependent and parameter-dependent Lyapunov function, sufficient conditions to guarantee the input-to-state stability (ISS) of the augmented fault estimation error system are established to meet the ISS-gain performance requirement. Then simultaneous estimation of the states and the faults is implemented using an observer-based estimator. The applicability of the developed fault estimation method is validated by numerical results.

Keywords: Fault detection/accomodation, Stochastic/uncertain systems, Observers
Abstract: The paper deals with the early detection of incipient faults in nonlinear continuous systems. The model-based diagnosis algorithm is based on a state-set observation. A state-set observer is a set-theoretic approach for systems with unknown but bounded uncertainties in the input and output measurements. The calculated predicted state-sets from the state-set observer are used to check the consistency with the current measurement sets. In this paper, we present an incipient fault detection algorithm based on an extended consistency test. The new consistency test calculates a trust factor with the measurement and predicted state-sets. A small value of the trust factor indicates a small match between the measurement sets and the predicted state-sets. The time behavior of the trust factor is used to detect incipient faults early on in a technical system. An early detection and warning of incipient faults is valuable for safer process operations and preventive protection of process equipment. Finally, the proposed incipient fault detection method is demonstrated using measured data of fault-free and faulty operation of a nonlinear water tank system.

Keywords: Fault detection/accomodation, Linear parameter-varying systems, LMIs
Abstract: This paper addresses the problem of distributed fault detection and isolation (FDI) observer design in finite frequency domain for a class of heterogeneous multi-agent linear parameter-varying (LPV) systems. For this purpose, each agent uses an unknown input observer (UIO) as the FDI module which generates the residual signal. The proposed distributed FDI problem is formulated as a multi-objective optimization problem based on finite frequency H_infty and H index performance indices to respectively measure disturbance robustness and fault sensitivity. Sufficient conditions for designing the FDI observers are obtained based on linear matrix inequality (LMI) conditions. It is shown that based on the constructed residual signals each agent is capable of detecting and isolating its own faults and also its neighbors’ faults. Finally, a numerical example is provided to demonstrate the applicability of the proposed design framework and its advantages over other approaches.

Keywords: Automotive applications

Keywords: Automotive applications

Keywords: Automotive applications

Keywords: Automotive applications

Keywords: Automotive applications

Keywords: Process control, Optimization, Predictive control
Abstract: We apply optimal control to operational optimization to a model of a real oil reservoir, SOLSORT, located in the Danish North Sea. The oil reservoir model is a partial differential equation system that is simulated using a commercial oil reservoir simulator (Eclipse 300). The numerical optimization is conducted using an optimization software tool called RESOPT that is implemented in Matlab. The net present value of the reservoir model is maximized using gradient-based optimization methods. Both linear constraints on the manipulated variables and nonlinear output constraints on gas production rates are considered. We impose the nonlinear output constraints using a soft constrained penalty function. Compared to a reference case based on best practice reservoir engineering, the solution based on optimal control is able to increase the net present value by 6-13% and constrain the gas production rate.

Keywords: Process control, Filtering, Nonlinear systems
Abstract: We investigate two algorithms for state estimation of nonlinear chromatographic processes: the Extended Kalman Filter and the Ensemble Kalman Filter. We consider the Equilibrium Dispersive Model for modeling of packed-bed chromatography. The Equilibrium Dispersive Model is governed by a convection dominated nonlinear partial differential equation. The model is discretized by a high-order discontinuous-Galerkin finite-element method for accurate and efficient simulation of the chromatographic process. The discretization leads to a stochastic continuous-discrete model, and we use the Extended Kalman Filter and the Ensemble Kalman Filter for state estimation of the packed-bed chromatographic model. The performance and capabilities of both filters are demonstrated in simultaneous estimation of the unknown system states and uncertain inlet concentration.

Keywords: Process control, Reduced order modeling, Identification
Abstract: To facilitate data-driven tuning of kalman filter, we propose a control-oriented greybox noise structure of multi-stage spray dryers. In the first step, we derive the noise structure by a combination of physical arguments, experimental data and control-oriented considerations. In the second step, we evaluate the proposed noise structure by comparison to a benchmark noise structure used in the industry. The comparison is based on experimental data from a pilot plant. The evaluations show that the proposed noise structure gives a difference only in noise-performance trade-offs. Hence the proposed noise structure is suitable to be used in closed loop control of multi-stage spray dryers.

Keywords: Reinforcement learning, Smart structures, Machine learning
Abstract: Mixing Loops are often used for proper pressurization and temperature control in building thermal systems. Optimal control of the mixing loop maximizes comfort while minimizing cost. To ensure optimal control for mixing loops in a wide range of different buildings with different load conditions, a self learning controller is here proposed. The controller uses Reinforcement Learning with flow variable eligibility trace. The controller is shown to improve performance of the mixing loop control compared to state of the art reinforcement learning and industrial grade controllers. The controller is tested on a hardware in the loop setup for rapid testing of mixing loop control used in building heating.

Keywords: Process control, Control applications, Nonlinear systems
Abstract: We evaluate the existing trade-off between the exhaust flow rates and temperatures on a air-cooled Open Compute Windmill V2 server platform. To this aim, we implement a novel flow provisioning controller that optimizes the server's exhaust air temperature, and compare its performance against that of the default provisioning strategy. We report both experimental and numerical results that link the attainable flow rates and the exhaust temperatures. We demonstrate thus the benefits of tailored control strategies in a heat recovery context, and provide crucial characterizations toward the investigation of data center heat recovery applications.

Keywords: Reduced order modeling, Simulation, Optimization
Abstract: Model order approximation of an incommensurate fractional order system is a highly challenging problem by any analytical technique. In view of this, a simple and a computationally efficient approach is proposed for the model order reduction of incommensurate fractional order systems. The proposed approach is based on FOMCON toolbox of MATLAB and hybrid Big bang Big crunch (HBBBC) optimization algorithm. The main advantage of the proposed technique is that Oustaloup approximation is not explicitly needed to approximate an incommensurate fractional order system into an equivalent higher integer order system, instead one could directly apply the proposed scheme on the fractional order transfer function. The proposed method is substantiated by two numerical examples from literature and an example of a pressurised heavy water reactor is also taken up. The simulation results are a testimony to the efficiency and efficacy of the proposed approach.

Keywords: Control applications, Biologically-inspired methods, Optimization
Abstract: An alternative Proportional-Integral-Derivative (PID) controller with two degrees of freedom will be proposed to control the translational position of quadrotor in this paper. This type of controller is able to deal with quadrotor complexity quite well. A new metaheuristic algorithm, Stochastic Fractal Search (SFS), is proposed to optimize the two degree of freedom PID controller parameters to ensure an optimal control action. For comparison purposes, the Particle Swarm Optimization (PSO) algorithm will also be implemented to optimize the controller parameters. The results show the superiority of SFS algorithm as an optimization algorithm, especially in this particular control scheme.

Keywords: PID control, Time delays, Genetic algorithms
Abstract: This paper studies the delay robustness problem for tracking control of second-order systems subject to unknown input delay and uncertain nonlinearity under PID control. The basic issue is to find the largest allowable input delay under PID control can achieve robust tracking in spite of the uncertain nonlinearity. First, the uncertain nonlinear delay system under PID control can be transferred into a delay system with time-varying uncertainty. Second, we develop the largest allowable time delay under fixed PID controller parameters by solving the generalized eigenvalue problem (GEVP). Third, a genetic algorithm (GA) is proposed to achieve the largest allowable delay and the corresponding PID parameters, which consists with the solutions under GEVP. The first-order uncertain nonlinear delay systems under PI control is investigated as well. Finally, two examples are given to show the effectiveness of the proposed method.

Keywords: Optimization, Aerospace applications, Planning
Abstract: In this paper, we present a numerical method for two-dimensional terrain following (TF) optimal trajectory regeneration. This method corrects the planned optimal trajectory to avoid collision with previously unknown obstacles. Firstly, an optimal TF trajectory is acquired using Hp-adaptive Gaussian quadrature collocation method. Then, a neighboring optimum feedback law is presented for trajectory regeneration using a backward sweep method. Numerical simulations are provided to validate the effectiveness and demonstrate the features of the problem.

Keywords: Distributed parameter systems, Identification, Reduced order modeling
Abstract: A scheme for fast kick estimation in managed pressure drilling is presented. The reservoir pressure and production index are estimated by utilizing information in a fast traveling shock wave induced by an unexpected rise in reservoir pressure. A least squares estimation problem is formulated from an early-lumping approach based on a reduced order drift-flux model. The Levenberg-Marquardt scheme is applied to solve the non-linear least-squares problem. The estimation scheme is tested with simulated top-side flow measurements from a two-phase drift flux model.

Keywords: Sliding mode control, Genetic algorithms, Autonomous systems
Abstract: The robust control of quadrotors has become increasingly important as their commercial use continues to increase. In this paper an optimal tuning mechanism for a sliding mode controller (SMC) used to control a quadrotor is proposed. The aim is to develop techniques for optimally tracking the vertical movement, attitude and heading of the quadrotor helicopter. An ant colony optimization (ACO) algorithm is used to tune eight SMC parameters. The performance of ACO is compared to the manually-tuned and genetic algorithm (GA)-tuned SMC. Through numerical simulations we show that ACO-SMC performs within reach of the GA-SMC and better than the manual tuning. The high-frequency chattering is solved by employing a saturation function. From the simulated results we conclude that, overall, the application of GA and ACO can be used to tune SMC parameters improves the quadrotor's performance.