This article provides an external stakeholder perspective on the influence of higher education in Sweden, exploring their views on curriculum development and qualitywork at the programme level. Semi-structured interviews with a selected number of representatives of external stakeholders involved in various educational areas were conducted at seven higher education institutions. The participants argued that changes intheir business sectors, and subsequent changes in the knowledge and skills in the labour needed, should encourage higher education institutions to adjust and develop their programmes. They did not anticipate or demand immediate changes in response to their comments, nor did they see themselves as a part of any quality assurance scheme. Uncertainties about the internal decision-making process and organisation in higher education institutions apparently do not facilitate external stakeholders’ understanding of their role in the larger scheme. However, all informants had comments on quality in higher education, perceiving it predominantly as something connected to the world of work. The practical implication of this study is that curriculum development at higher education institutions would benefit from communicating the internal decision-making processes to external stakeholders and agreeing on the expectations with them, in collaboration.

Modeling and failure prediction are important tasks in many engineering systems. For these tasks, the machine learning literature presents a large variety of models such as classification trees, random forest, artificial neural networks, among others. Standard statistical models such as the logistic regression, linear discriminant analysis, k-nearest neighbors, among others, can be applied. This work evaluates advantages and limitations of statistical and machine learning methods to predict failures in industrial robots. The work is based on data from more than five thousand robots in industrial use. Furthermore, a new approach combining standard statistical and machine learning models, named hybrid gradient boosting, is proposed. Results show that the hybrid gradient boosting achieves significant improvement as compared to statistical and machine learning methods. Furthermore, local joint information has been identified as the main driver for failure detection, whereas failure classification can be improved using additional information from different joints and hybrid models. (C) 2019 Elsevier Ltd. All rights reserved.

A platform for sensor fusion consisting of a standard smartphone equipped with the specially developed Sensor Fusion appis presented. The platform enables real-time streaming of data over WiFi to a computer where signal processingalgorithms, e.g., the Kalman filter, can be developed and executed in a Matlab framework. The platform is an excellenttool for educational purposes and enables learning activities where methods based on advanced theory can be implementedand evaluated at low cost. The article describes the app and a laboratory exercise developed around these new technologicalpossibilities. The laboratory session is part of a course in sensor fusion, a signal processing continuation course focused onmultiple sensor signal applications, where the goal is to give the students hands on experience of the subject. This is done byestimating the orientation of the smartphone, which can be easily visualized and also compared to the built-in filters in thesmartphone. The filter can accept any combination of sensor data from accelerometers, gyroscopes, and magnetometers toexemplify their importance. This way different tunings and tricks of important methods are easily demonstrated andevaluated on-line. The presented framework facilitates this in a way previously impossible.

This paper presents a data-driven approach to diagnostics of systems that operate in a repetitive manner. Considering that data batches collected from a repetitive operation will be similar unless in the presence of an abnormality, a condition change is inferred by comparing the monitored data against an available nominal batch. The method proposed considers the comparison of data in the distribution domain, which reveals information of the data amplitude. This is achieved with the use of kernel density estimates and the Kullback–Leibler distance. To decrease sensitivity to disturbances while increasing sensitivity to faults, the use of a weighting vector is suggested which is chosen based on a labeled dataset. The framework is simple to implement and can be used without process interruption, in a batch manner. The approach is demonstrated with successful experimental and simulation applications to wear diagnostics in an industrial robot gearbox and for diagnostics of gear faults in a rotating machine.

A Design-Build-Test (DBT) course in electronics is presented. The course is designed based on the CDIO (Conceive-Design-Implement-Operate) framework for engineering education. It is part of the curriculum of two engineering programs at Linköping University, Sweden, where it has been given successfully for a number of years. The cornerstones of the course consist of carefully designed learning outcomes based on the CDIO Syllabus, a structured project management model such that the project tasks are carried out according to professional and industry-like routines, with well-designed organisation of the staff supporting the course, and challenging project tasks.

A framework for iterative learning control (ILC) is proposed for the situation when an ILC algorithm uses an estimate of the controlled variable obtained from an observer-based estimation procedure. Assuming that the ILC input converges to a bounded signal, a general expression for the asymptotic error of the controlled variable is given. The asymptotic error is exemplified by an ILC algorithm applied to a flexible two-mass model of a robot joint.

A method for the detection and estimation of nonlinear distortions when identifying multivariable frequency response functions (FRF) is considered. The method is successfully applied to experimental data, which were collected in closed loop, from an industrial robot. The results show that nonlinear distortions are indeed present and cause larger variability in the FRF than the measurement-noise contributions.

Iterative learning control (ILC) is applied to a robot arm with joint flexibility. The ILC algorithm uses an estimate of the arm angle, where the estimate is computed using measurements of the motor angle and the arm angular acceleration. The design of the ILC algorithm is evaluated experimentally on a laboratory scale robot arm with good results.

In recent years a new class of soft real-time applications operating in unpredictable environments has emerged. Typical for these applications is that neither the resource requirements nor the arrival rates of service requests are known or available a priori. It has been shown that feedback control is very effective to support the specified performance of dynamic systems that are both resource insufficient and exhibit unpredictable workloads. To efficiently use feedback control scheduling it is necessary to have a model that adequately describes the behavior of the system. In this paper we experimentally evaluate the accuracy of four linear time-invariant models used in the design of feedback controllers. We introduce a model (DYN) that captures additional system dynamics, which a previously published model (STA) fails to include. The accuracy of the models are evaluated by validating the models with regard to measured data from the controlled system and through a set of experiments where we evaluate the performance of a set of feedback control schedulers tuned using these models. From our evaluations we conclude that second order models (e.g., DYN) are more accurate than first order models (e.g. STA). Further we show that controllers tuned using second order models perform better than controllers tuned using first order models.

In this article, the Conceive – Design – Implement – Operate (CDIO) Syllabus and CDIO Standards are introduced, and the question is posed as to whether or not national circumstances affect the ability of engineering programmes to meet CDIOrequirements. In particular, the extent to which representative programmes from the USA, Canada, Sweden and the UK cover the CDIO Syllabus is assessed and conclusions are drawn. The international applicability of the CDIO Syllabus also depends on the absence of conflict between the syllabus and national accreditation criteria. Based on the countries considered, the authors suggest that no conflict exists. Furthermore, it is argued that the CDIO Syllabus is aspirational and, as such, it complements the threshold requirements of national accreditation criteria. It is also argued that the CDIO Syllabus, coupled with the CDIO Standards, could form the basis for an international benchmark that would co-exist with any future global accreditation criteria and would serve to continuously improve engineering education.

In the Conceive – Design – Implement - Operate (CDIO) project course in automatic control, an autonomous unmanned aerial vehicle (UAV) is constructed, utilising an existing radio controlled model aircraft. By adding an inertial sensor that measures acceleration and rotation, together with a Global Positioning System (GPS) sensor, the aim is to construct an accurate positioning system. This is used by an onboard computer to calculate control surface signals to a set of servos in order to follow a predefined way-point trajectory. The project involves 17 students and comprises both positioning, control and hardware design. The main pedagogical goal is for students to apply their theoretical knowledge within a project framework in order to improve important aspects of their engineering skills in a realistic manner.

Recursive identification of a physically parameterized model of a single link flexible robot arm is considered. The parameters are estimated using a recursive prediction error method applied to a linear continuous time model structure and discrete time data. The algorithm is applied to real data from an industrial robot, and three important parameters are identified using only measurements of the motor angle. The experiments show that the particular parameter that is estimated will have big influence on the algorithm behavior.

Closed-loop identification of an industrial robot of the type ABB IRB 1400 is considered. Data are collected when the robot is subject to feedback control and moving around axis one. Both black-box and physically parameterized models are identified. A main purpose is to model the mechanical flexibilities. It is found that a model consisting of three-masses connected by springs and dampers gives a good description of the dynamics of the robot.

The convergence properties of iterative learning control (ILC) algorithms are considered. The analysis is carried out in a framework using linear iterative systems, which enables several results from the theory of linear systems to be applied. This makes it possible to analyse both first-order and high-order ILC algorithms in both the time and frequency domains. The time and frequency domain results can also be tied together in a clear way. Results are also given for the iteration-variant case, i.e. when the dynamics of the system to be controlled or the ILC algorithm itself changes from iteration to iteration.

Disturbance aspects of iterative learning control (ILC) are considered. By using a linear framework it is possible to investigate the influence of the disturbances in the frequency domain. The effects of the design filters in the ILC algorithm on the disturbance properties can then be analyzed. The analysis is supported by simulations and experiments.

The problem of tracking time-varying properties of a signal is studied. The somewhat contradictory notion of “time-varying spectrum” and how to estimate the “current” spectrum in an on-line fashion is discussed. The traditional concepts and relations between time and frequency resolution are crucial for this problem. We introduce two definitions for the time resolution of filters, essentially measuring the effective number of past data that are used to form the estimate. In, for example, wavelet transform techniques, frequency-dependent time resolutions are used so that fewer data are used at higher frequencies, thus enabling faster tracking of high-frequency components (at the price of worse frequency resolution). The main contribution of the paper is to show how this same feature can be introduced when estimating spectra via a time-varying, autoregressive model of the signal. This is achieved by a special choice of nominal covariance matrix for the underlying parameter changes.

This note deals with the performance of the recursive least squares algorithm when it is applied to problems where the measured signal is corrupted by bounded noise. Using ideas from bounding ellipsoid algorithms we derive an asymptotic expression for the bound on the uncertainty of the parameter estimate for a simple choice of design variables. This bound is also transformed to a bound on the uncertainty of the transfer function estimate.

Recursive identification of time varying systems using ARX models is considered, with focus on the accuracy of the transfer function estimates. Three recursive identification algorithms are studied, the least mean squares algorithm, the recursive least squares algorithm and the Kalman filter respectively. The model accuracy is studied in terms of algorithm design variables, input and disturbance signal properties and variations of the true system. Using asymptotic methods approximate expressions for the model quality are derived. The derived expressions are validated by simulations.

To track the time-varying dynamics of a system or the time-varying properties of a signal is a fundamental problem in control and signal processing. Many approaches to derive such adaptation algorithms and to analyse their behaviour have been taken. This article gives a survey of basic techniques to derive and analyse algorithms for tracking time-varying systems. Special attention is paid to the study of how different assumptions about the true system's variations affect the algorithm. Several explicit and semi-explicit expressions for the mean square error are derived, which clearly demonstrate the character of the trade-off between tracking ability and noise rejection.

A case study of the use of reflections within the Applied physics and electrical engineering program at Linköping University is presented. Reflections have been used for several years and they are done at four stages in the program, in terms of reflections at the end of the Introductory course in year one, design-implement experiences in year three and five, and a reflection document that is the last component of the Master’s thesis. In the first three stages a project model is used to support the planning and execution of the project, and in the project model the project work ends with a reflection. In the reflection document connected to the Master’s thesis the student reflects upon both the thesis work itself and the entire education program, according to the sections and subsections of the CDIO Syllabus. The paper describes how the reflections are integrated in the program. Experiences from student perspective are collected in a small-scale study via interviews with students from year one and year five. 

Challenge-based learning (CBL) is a learning approach that has received increased attention during  the last years. In some of the publications about CBL the connections to the CDIO (conceive-design-implement-operate) framework is discussed, and the purpose of this paper is to discuss these connections further. CBL has connections to both Problem-based learning (PBL) and Project-based learning (PjBL), but while these are learning approaches, the CDIO framework has a program perspective, including aspects of learning approaches, and hence a wider scope. The connections and differences between CBL and CDIO are discussed based on the components of the CDIO Standards.  

The CDIO Initiative is going through a process of reconsidering and updating the CDIO approach for engineering education development. Previous work resulted in substantial updates of the twelve CDIO standards and the introduction of "optionel" standards. This paper reports on a similar review and update of the CDIO Syllabus to version 3.0. It has been developed by a working group consisting of four sub-groups and iterated and refined guidedby feedback from the whole CDIO community. There are mainly three external drivers that motivate the changes: sustainability, digitalization, and acceleration. There is also an internal driver in the form of lessons learned within the CDIO community, from using the Syllabus in curriculum and course development. Approximately 70 updates are proposed, amongst them three additions on the X.X level, namely 1.4 Knowledge of Social Sciences and Humanities,3.1 Teamwork and Collaboration, and 5.3 Research.

The use of the CDIO framework in the development of the ECIU University is presented. The paper discusses the relatively moderate adaptations and modifications of the CDIO Syllabus and Standards that are necessary to make the documents applicable also in this context. Since challenge-based learning (CBL) is central learning format in the ECIU University, special attention is given to the connections between CBL method, the conceive-design-implement-operate sequence and project-based learning, which is central in the CDIO framework. The paper presents both general aspects and examples of the applications and activities within ECIU University and Linköping University (LiU). The main messages of the paper are that the development of the ECIU University will benefit from applying the CDIO framework since it offers references for what an education should give, in terms of knowledge and skills, and how an education program should be designed. In addition, the components of the CDIO framework require a moderate amount of adaptation to be directly applicable. Examples of the ongoing implementation activities at LiU. 

Introduction 

Higher education institutions (HEIs) face challenges assessing the relevance of educational programmes. Upon graduation, the student should have acquired knowledge and understanding, competence and skills as well as good judgement and approaches to operate in a changing labour market. Ideas on new programmes and courses mainly emanate from research findings identified at the HEIs. Needs and expectations from external stakeholders have the potential to further contribute if room for collaboration is created. Extensive rapid societal changes increase this need for collaboration. 

The Standards and Guidelines for Quality Assurance in the European Higher Education Area (ESG) state that higher education aims to fulfil multiple purposes, including preparing students for active citizenship, future careers, personal development and create a broad, advanced knowledge base and stimulate research and innovation (ENQA, 2015). However, different stakeholders may have other priorities. Therefore, quality assurance needs to take these different perspectives into account. 

Relevance is considered an aspect of quality in higher education, but many HEIs in Sweden lack structures, processes, and methods for assessing relevance and involving stakeholders in these processes.

This project aimed to increase the knowledge and provide methods to systematically assess relevance and use university-industry collaborations as tools for educational development. 

Methods/Approach 

This paper is a summary of the project MERUT focused on methods for assessment of relevance in higher education. The project was carried out during 2017-2020 with financial support from Vinnova (The Swedish Innovation Agency) and involved seven Swedish HEIs. The connection to future career paths is often stated as the primary factor to describe the relevance of educational programmes and was selected as the focus of MERUT. Data have been collected using workshops, meetings, literature reviews, interview studies, and surveys. Important parts of the work have been interviews with external stakeholders in different labour-market areas who, in various ways, are involved in higher education, most often in advisory boards. Also, interviews with quality coordinators at university programmes as well as at faculty and university management levels at each HEI have been carried out.

The seven participating Swedish universities in MERUT have been Karolinska Institutet, Kristianstad University, KTH Royal Institute of Technology, Mälardalen University, Linköping University, Stockholm University, and Umeå University.

Results The project resulted in knowledge, tools, and methods to work systematically with relevance in higher education. The results can be summarized as follows:

·         Interviews with HEIs showed that they collaborate with external stakeholders in many ways, primarily around teaching and learning and to a lesser extent around programme management and quality assurance. 

·         Further, the involvement of stakeholders varied both between and within the universities (faculties, subject areas, levels of education). Therefore, it is difficult to evaluate, in a systematic way, how collaboration is included in the quality systems of the HEIs. 

·         A toolbox with methods for how to involve external stakeholders in the process of assessing and developing the relevance of a study programme. These methods can be used in continuous quality work, in major curriculum revisions as well as in the establishment of new programmes.

·         A checklist for external stakeholders' involvement in educational development to facilitate and clarify roles, structures, and tasks in connection with external stakeholders in both the development and operational phases of a study programme. 

The results show that there are many similarities between the HEIs in the study in terms of relevance assessment and dimensioning decisions. However, the potential of a systematic collaboration with stakeholders and society for relevance assessment and dimensioning of education is not yet fully being realised. 

Conclusion 

HEIs interact with external stakeholders in many ways within education. However, it is rare to find examples where external stakeholders are involved in the quality assurance process, at least not in a systematic way. The MERUT project has developed recommendations for collaboration perspectives and stakeholder participation in the governing of educational programmes. A systematic dialogue and interaction with stakeholders contribute to a mutual understanding of different stakeholder groups’ needs and expectations, and their view of quality of higher education. Furthermore, to consider relevance as a quality aspect creates a basis for a more methodical assessment process where external stakeholders can contribute in a clear role. MERUT has developed a toolbox and a checklist to facilitate such systematic interaction and collaboration with stakeholder groups. A conclusion of  this project is that a reciprocal, transparent, and systematic approach leads to a sustainable educational collaboration with improved quality and relevance of higher education.

It would constitute a large gain for society if the HEIs are able to systematically and by efficient processes take external stakeholders’ and societal needs and expectations into account when building comprehensive and systematic relevance assessment processes and in dimensioning of education.  

References

European Association for Quality Assurance in Higher Education (ENQA). (2015). Standards and Guidelines for Quality Assurance in the European Higher Education Area (ESG). Brussels, EURASHE.

An iterative, learning based, feed-forward method for compensation offriction in industrial robots is studied. The method is put into an ILC framework by using a two step procedure proposed inliterature. The friction compensation method is based on ablack-box friction model which is learned from operational data,and this can be seen as the first step in the method. In the second step the learned model is usedfor compensation of the friction using the reference joint velocityas input. The approach is supported by simulation experiments.

The CDIO Syllabus survey has successfully been applied to the Bachelor’s and Master’s programs in Experimental and Medical Biosciences, within the Faculty of Medicine and Health Sciences at Linköping University, Sweden. The programs are and have been, subject to considerable redesign with strong influence from the CDIO framework. One of the main drivers for the redesign is a shift concerning the main job market after graduation, from an academic career to industry and healthcare. One of the steps in the development process has been to carry out a Syllabus survey based on an adapted version of the CDIO Syllabus. The survey was sent out to students and to various categories of professionals, and in total 87 responses were received. The adapted version of the Syllabus and the design, execution, and outcome of the survey is presented.

The CDIO framework is an integrated and important part of the new quality assurance system within the Faculty of Science and Engineering at Linköping University. Both the CDIO Syllabus and the CDIO Standards are used extensively in the system. First, the paper presents the development and use of the second generation of course matrices (previously denoted ITU-matrices) and program matrices, which build upon an adapted and extended version of the CDIO Syllabus. The extension is made to also include bachelor’s and master’s program in subjects outside the engineering field. Second, the paper presents how the CDIO Standards are used in the quality reports, which are vital parts of the quality assurance systems. As a result, the CDIO framework is used for the design, management, and quality assurance of all education programs ( approximately 60 programs) within the Faculty of Science and Engineering at Linköping University.

A conceive-design project is presented. The project runs over the first two years of the engineering education programs at the Methodist University of São Paulo, and the paper focuses on the first semester of the Computer Engineering Program. The projects are carried out in teams of students from different programs and the results are presented to a board of faculty members and sometimes participants from our partner companies, at the end of each semester. There are around 200 students involved in total. For the first semester, students are required to conceive, design and document an environmentally-friendly product or service. The conceive-design project is integrated with surrounding modules is the curriculum, such as Entrepreneurship and Innovation and Economics, Society and Environmental Issues.

The CDIO framework for development of engineering education is presented, including the overall ideas, the fundamental documents, and some development tools. The automatic control subject and its role in engineering education is studied using the CDIO Standards as reference. Some examples from the engineering education at Linkoping University are presented with special focus on the control education. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

This paper describes the motivation, the current state and the further actions of an improvement process of the engineering education at the Military Institute of Engineering (IME) in Brazil. Based on the reasons for why and how to change, the CDIO framework has been chosen as the kernel of this improvement process. The activities realized the plan of the further actions and the open questions are presented in this paper. The paper is a condensed presentation of the report (Cerqueira et. al., 2016), where a thorough background and more details can be found.

Fault detection algorithms (FDAs) process data to generate a test quantity. Test quantities are used to determine presence of a fault in a monitored system, despite disturbances. Because only limited knowledge of the system can be embedded in an FDA, it is important to evaluate it in scenarios relevant in practice. In this paper, simulation based approaches are proposed in an attempt to determine: i) which disturbances affect the output of an FDA the most; ii) how to compare the performance of dierent FDAs; and iii) which combinations of fault change size and disturbances variations are allowed to achieve satisfactory performance. The ideas presented are inspired by the literature of design of experiments, surrogate models, sensitivity analysis and change detection. The approaches are illustrated for the problem of wear diagnosis in manipulators where three FDAs are considered. The application study reveals that disturbances caused by variations in temperature and payload mass error affect the FDAs the most. It is also shown how the size of these disturbances delimit the capacity of an FDA to relate to wear changes. Further comparison of the FDAs reveal which performs "best" in average.

Three approaches of iterative learning control (ILC) applied to a Gantry-Tau parallel kinematic robot are studied; ILC algorithms using 1) measured motor angles, 2) tool-position estimates, and for evaluation purposes, 3) measured tool position. The approaches are compared experimentally, with the tool performance evaluated using external sensors. It is concluded that the tool performance can be improved using tool-position estimates in the ILC algorithm, compared to when using motor-angle measurements. Applying ILC algorithms to a system following trajectories with so-called lead-in/lead-out is also considered in the paper.

An example of how entrepreneurship can be integrated into Design-Build-Test (DBT) project courses is presented. The example is taken from the engineering program Applied Physics and Electrical Engineering at Linköping University, where entrepreneurship has been introduced in ten different DBT project courses related to the specializations of the program. The purpose of the entrepreneurship part is that the students shall acquire knowledge and abilities within the general area of entrepreneurship with particular focus on business planning for new ventures. The organization and execution of the entrepreneurship activities are described in detail together with a summary of the experiences from the first year. The results of the entrepreneurship activities within the project courses are positive and will be further developed, with emphasis on the connections between project ideas and technical contents of the courses.

A framework for Iterative Learning Control (ILC) is proposed for the situation when the ILC algorithm is based on an estimate of the controlled variable obtained from an observer-based estimation procedure. Under the assumption that the ILC input converges to a bounded signal, a general expression for the asymptotic error of the controlled variable is given. The asymptotic error is then exemplified by an ILC algorithm applied to a flexible two-mass model of a robot joint.

ILC is traditionally applied to systems where the controlled variable is the measured variable. However, in standard industrial robots only the motor angles are measured, while the control objective is to follow a tool path. A modern industrial robot is flexible, and assuming that the mechanical flexibilities are concentrated to the robot joints (elastic gearboxes), a flexible two-mass model can be used to describe a single joint. A P-ILC algorithm is applied to the two-mass model, based on only measured angle of the first mass (motor angle) or estimated angle for the second mass (tool angle). Robustness of the algorithm, and performance when model errors are introduced in the model, are discussed considering the error of the tool angle. First, it can be concluded for a flexible system that the characteristics of the motor-angle reference is essential for the resulting tool angle when the tool angle cannot be measured. Second, using an estimate of the tool angle instead of the explicit motor angle in the ILC update reduces the tool-angle error significantly.

This paper describes faculty professional competence development programs at Chalmers University of Technology, the Technical University of Denmark, and Linköping University. Examples of professional competences include project management, communication, teamwork and organizational change management. The description of the programs is complemented by interviews with faculty aiming at clarifying the needs for and experiences from faculty professional competences development programs.

When an ILC algorithm is applied to an industrial robot, the goal is to move the tool along a desired trajectory, while only the motor position can be measured. In this paper aspects of robustness and performance are discussed when an ILC algorithm is applied to a flexible two-mass system. It is shown that the stabilising controller of the two-mass system also directly affects the robustness properties of the ILC algorithm. A classical non-causal P-ILC algorithm and a model-based ILC design using optimisation are applied to the system, based on the error for the first mass. Performance and robustness of the algorithms are compared when model errors are introduced in the system, showing that the optimisation-based approach can handle larger model uncertainties. It is illustrated that the performance of the overall system, when considering position of the second mass, is the practical limit compared to the limiting factor of the robustness of the ILC algorithms.

A benchmark problem for robust feedback control of a manipulator is presented. The system to be controlled is an uncertain nonlinear two link manipulator with elastic gear transmissions. The gear transmission is described by nonlinear friction and elasticity. The system is uncertain according to a parametric uncertainty description and due to uncertain disturbances affecting both the motors and the tool. The system should be controlled by a discrete-time controller that optimizes performance for given robustness requirements. The control problem concerns only disturbance rejection. The proposed model is validated by experiments on a real industrial manipulator.

Experimental evaluation of an Iterative Learning Control (ILC) algorithm is presented. The ILC algorithm is applied to all motors of a large size commercial six degrees-of-freedom industrial robot in order to minimise the error measured on the motor angles. The performance of the algorithm is evaluated with respect to the operating point of the robot, the programmed tool velocity, and the design variables of the ILC algorithm. The chosen movements are intended to represent typical paths in a laser cutting application. Even though a fairly simple ILC algorithm is applied, the error reduction is substantial after only five iterations and the algorithm shows good robustness properties.

This paper deals with differentiated services in real-time systems. Tasks submitted to a real-time system are differentiated with respect to importance and QoS requirements. We use feedback control to enforce the requirements in QoS and ensure a hierarchical admission policy based on the importance of the tasks. The results show that the requirements are met during steady state when the workload is constant. The feedback control approach does not satisfactorily manage QoS when there is a sudden and significant workload change (transient state) due to the time-variant nature of the system. To address this, we present preliminary and promising results using adaptive control, and report on some challenges we are facing when applying the theory.

System identification in robotics often involves the estimation of linear models characterizing the behavior in certain operating points. In this paper, a method for the detection and estimation of nonlinear distortions in an estimated frequency response function (FRF) has successfully been applied to experimental data from an industrial robot. The results show that nonlinear distortions areindeed present and cause larger variability in the FRF than the measurement noise contributions.

A first step towards making a toolbox in Maple for industrial robot modelling is taken. Position and orientation of the tool can be determined in terms of the Denavit-Hartenberg joint variables and also the Jacobian relating the linear and angular velocities to the joint velocities. Further on it will be possible to, eg, differentiate the Jacobian. Future work includes to evaluate different kinds of sensors and sensor locations and symbolically generate the kinematic models using Maple. It also means to incorporate the models in Matlab- or C-code for including the results, eg, in an Extended Kalman Filter algorithm for state estimation.

Nonlinear grey-box identification of industrial robots is considered. A three-step identification procedure is proposed in which parameters for rigid body dynamics, friction and flexibilites can be identified only using measurements on the motor. In the first two steps, good initial parameter estimates are derived which are used in the last step, where the parameters of a nonlinear physically parameterized model are identified directly in the time domain. The procedure is examplifed using real data from an experimental industrial robot.

This paper presents a gain scheduling control of a nonlinear system in which the reference trajectory is given in advance. Multiple frozen operating times are chosen on the reference trajectory and a linear time invariant model is obtained at each operating time. A linear parameter varying model is then constructed by interpolating the region between the neighbor frozen operating times. A gain scheduling state feedback law is designed by a linear matrix inequality formulation. The effectiveness is demonstrated in a numerical simulation of a traing control of a two-link robot arm.

The intricacy of real-time data service management increases mainly due to the emergence of applications operating in open and unpredictable environments, increases in software complexity, and need for performance guarantees. In this paper we propose an approach for managing the quality of service of real-time databases that provide imprecise and differentiated services, and that operate in unpredictable environments. Transactions are classified into service classes according to their level of importance. Transactions within each service class are further classified into subclasses based on their quality of service requirements. In this way transactions are explicitly differentiated according to their importance and quality of service requests. The performance evaluation shows that during overloads the most important transactions are guaranteed to meet their deadlines and that reliable quality of service is provided even in the face of varying load and execution time estimation errors.

Iterative learning control (ILC) is applied to a laboratory scale robot arm with joint exibility. The ILC algorithm is based on an estimate of the arm angle, where the estimate is formed using measurements of the motor angle and the arm angular acceleration. The design of the ILC algorithm is based on a model obtained from system identication. The ILC algorithm is evaluated experimentally on the robot arm with good results.

Iterative learning control (ILC) is applied to a laboratory scale robot arm with joint flexibility. The ILC algorithm is based on an estimate of the arm angle, where the estimate is formed using measurements of the motor angle and the arm angular acceleration. The design of the ILC algorithm is based on a model obtained from system identification. The ILC algorithm is evaluated experimentally on the robot arm with good results.

Recursive identification of physically parameterized models of continuous time systems is considered. As an example a model of a single link flexible robot arm is considered. The aim of the identification is to generate online estimates of physical parameters that can be used for e.g. diagnosis purposes. For evaluation the algorithm is applied to data from an industrial robot, and three important parameters are identified using only measurements of the motor angle.

A simulation and animation tool for education in multivariable control is presented. The purpose of the tool is to support studies of various aspects of multivariable dynamical systems and design of multivariable feedback control systems. Different ways to use this kind of tool in control education are also presented and discussed.

Feed-back control of a flexible mechanical system is considered using a linear two-mass model. It is found that unstable controllers often occur when the performance requirements of the closed loop system are made too high. It is illustrated how potential windup problems caused by unstable controllers in combination with input saturations can be handled. A remaining consequence of the use of unstable controllers is that it puts stronger requirements on the accuracy of the model. This is illustrated using the Bode integral.

A synthesis algorithm for the filters in a first order ILC is presented and applied on an industrial robot. The proposed ILC synthesis method is evaluated using two experiments on the robot. The first is a one-axis experiment where the system can be seen as a single servo. A modeling experiment is done to give input to the synthesis algorithm and then ILC is applied to the single axis showing a dramatic improvement in trajectory following. In the second experiment ILC is applied to a more complex multi axes motion where the robot draws a circle in a plane. The evaluation of the result is done using a pen mounted on the robot and it is evident that also on the arm-side an improved motion can be achieved. In both experiments the error converges to a stable level in about 5 iterations. Since a model is desired for the synthesis, an extra iteration has to be done for the modeling experiment. In this particular case a good path following can therefore be achieved after 6 iterations.

A frequency domain analysis method of a second order iterative learning control (ILC) algorithm is considered. Using the notion of iterative systems bounds for stability are presented in the frequency domain for the second order term. The bounds are found using a geometrical approach based on the special structure of the transfer matrix in the iterative system. Two examples are included showing how the analysis method can be used in an application

LQG control of a flexible servo system is considered using a linear two-mass model. The control system is mainly based on the measured position before the flexibility, but the use of acceleration feedback is also studied. Some aspects of regulator stability are considered.

A frequency domain analysis of a second order Iterative Learning Control (ILC) algorithm is considered. It is shown that an unstable first order ILC can be stabilized by the introduction of a second order term in the ILC. Bounds for stability are presented in the frequency domain for the second order term. The bounds are found using a geometrical approach based on the special structure of the transfer matrix in the iterative system. Two examples are included showing that introducing the higher order ILC can both stabilize an unstable first order ILC algorithm, as well as give better convergence properties compared to using a first order ILC algorithm.

Iterative Learning Control is presented briefly together with some results on convergence and some new results on iterative learning control with measurement disturbances. In the example given, iterative learning control is used in combination with conventional feed-back and feed-forward control, and it is shown that the control is highly affected by the disturbances. Some ideas on how the filters in the algorithm should be chosen are also discussed.

Many manipulators at work in factories today repeat their motions over and over in cycles and if there are errors in following the trajectory these errors will also be repeated cycle after cycle. The basic idea behind iterative learning control (ILC) is that the controller should learn from previous cycles and perform better every cycle. Iterative learning control is used in combination with conventional feed-back and feed-forward control, and it is shown that learning control signal can handle the effects of unmodeled dynamics and friction. Convergence and disturbance effects as well as the choice of filters in the updating scheme are also addressed.

The combination of tracking and regularization in recursive identification is studied. It is shown that regularization of the information matrix corresponds to a normalization of the covariance matrix, and that several of the proposed methods for dealing with covariance matrix blow up can be interpreted as approximate implementations of covariance matrix normalization.

In this paper the relationships between covariance modification and regularization in recursive least squares identification are investigated. An update equation for the information matrix is derived and it is shown how regularization of the information matrix can be expressed as a particular type of covariance matrix modification. The paper also presents an analysis of the effects of a covariance modification for obtaining regularization that was proposed in Salgado et al. (1988)

In this contribution we propose an optimization approach to the design of a restricted complexity controller. The design criterion is of LQG type containing two terms. The first term is the quadratic norm of the error between the output of the true closed loop and a desired response. The second term is the quadratic norm of the input signal. It is shown that the minimization of this criterion does not require a model of the system. Closed loop experimental data can be used instead. The result is an iterative scheme of closed loop experiments and controller updates which converges to a local minimum of the design criterion under the condition of bounded signals.

In this paper we present an approach to adaptive control of hydraulic actuators with flexible mechanical loads. The approach is based on recursive identification of low order models of the dynamics from valve input to actuator position and effective load pressure respectively. The model parameters are used to compute a regulator that gives a well damped system with good servo properties. The identification is carried out on-line in closed loop with no other signals than those present in normal operation. Here RLS with variable forgetting factor is used. The method does not, however, need not more paramters to be set a priori than ordinary RLS with fix forgetting factor. In this way an easily tuned and robust identification algorithm is obtained. Results from experiments carried out on a lorry crane are presented.

In servo applications involving complex mechanical structures load forces often cause deformations in the structure. Since normally only the actuator position is measured and fed back to the control unit the deformations will introduce errors in the positioning of the load. The approach proposed in this paper is to estimate the true load position. If the stiffness between the actuator and the load is known it is possible to estimate the position of the load by measuring the actuator force/pressure. The estimation of the stiffness is based on recursive identification of the dynamics between actuator force and actuator pressure. The identified model can be used to improve the dynamic properties of the system as well as to compensate for the static error.

A study is made of three recursive identification algorithms with respect to their tracking capability and disturbance sensitivity. The algorithms are applied to the identification of time-varying systems with finite impulse response. The performance of the algorithms are measured in terms of the mean square error of the estimated transfer function. Using asymptotic methods, explicit expressions for the mean square error are derived, and these expressions are used to evaluate the performance of the algorithms when applied to systems with different behavior.

The frequency domain properties of pole placement regulators are studied, and the influence of the regulator on the model quality is discussed. Some methods to affect the regulator properties are discussed.

In this report we study the mean square error of transfer function estimates obtained by identification. The model quality is related to the use of models in regulator design, and the properties of affecting the bias part of the mean square error, both in fixed parameter control and adaptive control.

We also study the variance part of the mean square error for transfer function estimates obtained by recursive identification. In the report we derive expressions for the variance as a function of the step length of the identification method, the input spectrum, the variations of the true system and the model order. These expressions are verified and utilized in simulations.

This report describes the motivation, the current state and the future actions of an improvement process in engineering education at the Brazilian higher education institution called the Military Institute of Engineering. Based on the reasons for why and how to change, the CDIO framework was chosen, at the end of 2014, as the kernel of this improvement process. The activities realized, the plan for the future actions and the open questions are presented in this report.

Friction is the result of complex interactions between contacting surfaces in a nanoscale perspective. Depending on the application, the different models available are more or less suitable. Available static friction models are typically considered to be dependent only on relative speed of interacting surfaces. However, it is known that friction can be affected by other factors than speed. In this paper, the typical friction phenomena and models used in robotics are reviewed. It is shown how such models can be represented as a sum of linear and nonlinear functions of relevant states, and how the identification method described in [1] can be used to identify them when all state sare measured. The discussion follows with a detailed experimental study of friction in a robot joint under changes of joint angle, load torque and temperature. Justified by their significance, load torque and temperature are included in an extended static friction model. The proposed model is validated in a wide operating range, considerably improving the prediction performance compared to a standard model.

Some implementation aspects of Iterative Learning Control (ILC) are considered. Since the ILC algorithm involves filtering of various signals over finite time intervals, often using non-causal filters, it is important that the boundary effects of the filtering operations are handled in an appropriate way when implementing the ILC algorithm. The paper illustrates in both theoretical analysis using the matrix description and in simulations of a twomass system that the method of implementation for handling the boundary effects can have large influence over stability and convergence properties of the ILC algorithm.

Estimates from an extended Kalman filter (EKF) is used in an Iterative Learning Control (ILC) algorithm applied to a realistic two-link robot model with flexible joints. The angles seen from the arm side of the joints (arm angles) are estimated by an EKF in two ways: 1) using  measurements of angles seen from the motor side of the joints (motor angles), which normally  are the only measurements available in commercial industrial robot systems, 2) using both motor- angle and tool-acceleration measurements. The estimates are then used in an ILC algorithm. The results show that the actual arm angles are clearly improved compared to when only motor angles are used in the ILC update, even though model errors are introduced.

When an ILC algorithm is applied to an industrial robot, the goal is to move the tool along a desired trajectory, while only the motor position canbe measured. In this paper aspects of robustness and performance are discussed when an ILC algorithm is applied to a flexible two-mass system. It is shown that the stabilising controller of the two-mass system also directly affects the robustness properties of the ILC algorithm. A classical noncausa lP-ILC algorithm and a model-based ILC design using optimisation are applied to the system, based on the error for the first mass. Performance and robustness of the algorithms are compared when model errors are introduced in the system, showing that the optimisation-based approach can handle larger model uncertainties. It is illustrated that the performance of the overall system, when considering position of the second mass, is the practical limit compared to the limiting factor of the robustness of the ILC algorithms.

In this paper a step toward making a toolbox for industrial robot modelling based on the computer algebra tool Maple is taken. It can be seen as a modelling platform for efficient derivation of the necessary equations for doing, eg, sensor fusion or state estimation by an Extended Kalman Filter (EKF) algorithm. Forward kinematics is studied and the position and orientation of the robot tool are determined in terms of the Denavit-Hartenberg joint variables. Linear and angular velocities and accelerations are derived using the Jacobian. The toolbox is exemplified using an IRB1400 from ABB Robotics with a so called parallelogram linkage structure. The kinematic relations received are verified using the robot IRB1400 with an accelerometer placed on the robot tool and it is shown that measured acceleration and theoretical acceleration derived from the kinematics agree well. However no flexibilities in the modelling are taken into account, which results in differences between measured and derived acceleration when several motors of the robot cooperate in the motion.

Estimation methods for the multivariable frequency response function are analyzed, both in open and closed loop. Expressions for the bias and covariance are derived and the usefulness of these expressions is illustrated in simulations of an industrial robot where the different estimators are compared. The choice of estimator depends on the signal-to- noise ratio as well as the measurement setup and a bias-variance trade-off.

A first step towards making a toolbox in Maple for industrial robot modelling is taken. Position and orientation of the tool can be determined in terms of the Denavit-Hartenberg joint variables and also the Jacobian relating the linear and angular velocities to the joint velocities. Further on it will be possible to, eg, differentiate the Jacobian. Future work includes to evaluate different kinds of sensors and sensor locations and symbolically generate the kinematic models using Maple. It also means to incorporate the models in Matlab- or C-code for including the results, eg, in an Extended Kalman Filter algorithm for state estimation.

System identification in robotics often involves the estimation of linear models characterizing the behavior in certain operating points. In this paper, a method for the detection and estimation of nonlinear distortions in an estimated frequency response function (FRF) has successfully been applied to experimental data from an industrial robot. The results show that nonlinear distortions areindeed present and cause larger variability in the FRF than the measurement noise contributions.

Experimental evaluation of an Iterative Learning Control (ILC) algorithm is presented. The ILC algorithm is applied to all motors of a large size commercial six degrees-of-freedom industrial robot in order to minimise the error measured on the motor angles. The performance of the algorithm is evaluated with respect to the operating point of the robot, the programmed tool velocity, and the design variables of the ILC algorithm. The chosen movements are intended to represent typical paths in a laser cutting application. Even though a fairly simple ILC algorithm is applied, the error reduction is substantial after only five iterations and the algorithm shows good robustness properties.

Nonlinear grey-box identification of industrial robots is considered. A three-step identification procedure is proposed in which parameters for rigid body dynamics, friction and flexibilites can be identified only using measurements on the motor. In the first two steps, good initial parameter estimates are derived which are used in the last step, where the parameters of a nonlinear physically parameterized model are identified directly in the time domain. The procedure is examplifed using real data from an experimental industrial robot.

This paper presents a gain scheduling control of a nonlinear system in which the reference trajectory is given in advance. Multiple frozen operating times are chosen on the reference trajectory and a linear time invariant model is obtained at each operating time. A linear parameter varying model is then constructed by interpolating the region between the neighbor frozen operating times. A gain scheduling state feedback law is designed by a linear matrix inequality formulation. The effectiveness is demonstrated in a numerical simulation of a traing control of a two-link robot arm.

Some fundamental limitations of causal and Current Iteration Tracking Error (CITE) discrete time Iterative Learning Control (ILC) algorithms are stud- ied using time and frequency domain convergence criteria. Of particular interest are conditions for monotone convergence, and these are evaluated using a discrete- time version of Bode's integral theorem. A relation between the frequency domain convergence conditions and the time-domain monotone convergence criterion is also discussed.

Iterative learning control (ILC) is applied to a laboratory scale robot arm with joint flexibility. The ILC algorithm is based on an estimate of the arm angle, where the estimate is formed using measurements of the motor angle and the arm angular acceleration. The design of the ILC algorithm is based on a model obtained from system identification. The ILC algorithm is evaluated experimentally on the robot arm with good results.

Frequency domain convergence conditions for Current Iteration Tracking Error (CITE) Iterative Learning Control (ILC) algorithms are presented. The convergence conditions together with a discrete-time version of Bodes integral theorem imply a strong restriction upon the kind of systems that can be controlled using a CITE ILC algorithm. The restriction is caused by the fact that the filter in the ILC algorithm has to be causal and and part of a closed loop system that has to be stable.

Some initial results on the disturbance properties of high order iterative learning control (ILC) algorithms are presented. Of particular interest is to investigate how high order ILC algorithms cope with measurement and load disturbances of different character. Some results are obtained by assuming statistical properties of the involved disturbances and deriving equations for the covariance matrix of the control error vector. The results are illustrated by analytic derivation of the covariance matrix for a second order ILC algorithm with a particular choice of design variables.

Closed-loop identification of an industrial robot of the type ABB IRB 1400 is considered. Data are collected when the robot is subject to feedback control and moving around axis one. Both black-box and physically parameterized models are identified. A main purpose is to model the mechanical flexibilities. It is found that a model consisting of three-masses connected by springs and dampers gives a good description of the dynamics of the robot.

Disturbance aspects of iterative learning control (ILC) are considered. By using a linear framework it is possible to investigate the influence of the disturbances in the frequency domain. The effects of the design filters in the ILC algorithm on the disturbance properties can then be analyzed. The analysis is supported by simulations and experiments.

Disturbance aspects of iterative learning control (ILC) are considered. By using a linear framework it is possible to investigate the influence of the disturbances in the frequency domain. The effects of the design filters in the ILC algorithm on the disturbance properties can then be analyzed. The analysis is supported by simulations and experiments.

An optimization approach to Iterative Learning Control (ILC) is considered. The ILC algorithm is formed by minimizing a quadratic criterion in the control error and input signal. A frequency domain interpretation of the derived updating algorithm is given. Experiments carried out on an ABB IRB 1400 are presented.

A synthesis algorithm for the filters in a first order ILC is presented and applied on an industrial robot. The proposed ILC synthesis method is evaluated using two experiments on the robot. The first is a one-axis experiment where the system can be seen as a single servo. A modeling experiment is done to give input to the synthesis algorithm and then ILC is applied to the single axis showing a dramatic improvement in trajectory following. In the second experiment ILC is applied to a more complex multi axes motion where the robot draws a circle in a plane. The evaluation of the result is done using a pen mounted on the robot and it is evident that also on the arm-side an improved motion can be achieved. In both experiments the error converges to a stable level in about 5 iterations. Since a model is desired for the synthesis, an extra iteration has to be done for the modeling experiment. In this particular case a good path following can therefore be achieved after 6 iterations.

Tuning of robot joint controllers using iterative feedback tuning (IFT) is considered. Using a simulation model of a two-link robot arm the applicability of IFT is investigated for a nonlinear and multivariable system. Due to the strong crosscoupling between the two links a decoupling controller is needed in order to obtain satisfactory results. An approach for obtaining a decoupling controller is investigated. It is also shown how IFT can be used to tune the controller parameters for a particular trajectory of the arm.

An introduction to Iterative Learning Control (ILC) is given. The basic principle behind ILC in both open loop and closed loop problems is explained. A general class of algorithms for updating of the ILC input signal is presented and the choice of the filters in the update algortihm is discussed with respect to convergence, robustness and disturbance sensitivity.

Many manipulators at work in factories today repeat their motions over and over in cycles and if there are errors in following the trajectory these errors will also be repeated cycle after cycle. The basic idea behind iterative learning control (ILC) is that the controller should learn from previous cycles and perform better every cycle. Iterative learning control is used in combination with conventional feed-back and feed-forward control, and it is shown that learning control signal can handle the effects of unmodeled dynamics and friction. Convergence and disturbance effects as well as the choice of filters in the updating scheme are also addressed.

Control of a hydraulic crane is considered. Due to the oscillatory character of the system smooth the operation of the crane is a demanding task. In order to improve the handling properties feedback control of the crane is studied. Based on linearized models feedback regulators of both LQG and PID type are designed. The feedback is based on position, pressure and acceleration measurements. Since the properties of the system change with load and operating point adaptive control is also introduced. The use of accelerometer signals for impact detection is also discussed. The proposed solutions are tested in both simulations and experiments on a real crane.

Issues concerning diagnosis, supervision and saftey are found in many technologically advanced products. There is now a trend to extend the functionality of diagnosis and supervision systems to handle more advanced situations. This report collects some of the initiatives taking place in research and some of the developments taking place in the industry.

We demonstrate that some recently proposed itera-tive identification and control design schemes do not necessarily converge to a local minimum of the design objective in the case of a restricted complexity model. There is, however, a link between these approaches and a recently proposed iterative optimization based control design procedure based on experimental data. We show that if the achieved and the desired output responses are perfectly matched, the schemes are (essentially) equivalent under noise free conditions.

In this report we study some different aspects of schemes for iterative identification and control design. By formulating the control objective as a criterion minimization task the problem of finding a model well suited for control appears to be closely related to a prediction error minimization problem in system identification. We discuss two ways of matching the control and identification criteria and evaluate their properties in simulations.

This paper presents a control system based on linear quadratic state feedback and state estimation for an hydraulic actuator with a flexible mechanical load. The purpose of the control system is to improve the dynamic properties by reducing the oscillatory behavior of the load and to eliminate the steady state error in the load position caused by external disturbances. Using a simplified linear model a control system is developed and evaluated using simulations and experiments.

This note deals with the performance of the recursive least squares algorithm when it is applied to problems where the measured signal is corrupted by bounded noise. Using ideas from bounding ellipsoid algorithms we derive an asymptotic expression for the bound on the uncertainty of the parameter estimate for a simple choice of design variables. This bound is also transformed to a bound on the uncertainty of the transfer function estimate.

Recursive identification of time varying systems with ARX structures is considered, with the interest focused on the quality of the estimated transfer functions. Three recursive identification algorithms are considered, the least mean squares (LMS) algorithm, the recursive least squares (RLS) algorithm and the Kalman filter respectively. The model quality is studied in terms of algorithm design variables, input and disturbance signal properties and variations of the true system. Using asymptotic methods approximate expressions for the model quality are derived.

Frequency domain expressions for the quality of recursively identified Laguerre models are presented. These models generalize finite impulse response models by using a priori information about the dominating time constants of the system to be identified. Expressions for the model quality are derived under the assumptions that the system varies slowly, that the model is updated slowly, and that the model order is high. The model quality is evaluated by investigating the properties of the estimated transfer function, and explicit expressions for the mean square error of the transfer function estimate are derived.

A study is made of three recursive identification algorithms with respect to their tracking capability and disturbance sensitivity. The algorithms are applied to the identification of time-varying systems with finite impulse response. The performance of the algorithms are measured in terms of the mean square error of the estimated transfer function. Using asymptotic methods, explicit expressions for the mean square error are derived, and these expressions are used to evaluate the performance of the algorithms when applied to systems with different behavior.

In servo applications involving complex mechanical structures load forces often cause deformations in the structure. Since normally only the actuator position is measured and fed back to the control unit the deformations will introduce errors in the positioning of the load. The approach proposed in this paper is to estimate the true load position. If the stiffness between the actuator and the load is known it is possible to estimate the position of the load by measuring the actuator force/pressure. The estimation of the stiffness is based on recursive identification of the dynamics between actuator force and actuator pressure. The identified model can be used to improve the dynamic properties of the system as well as to compensate for the static error.

The variance of recursive transfer function estimates for systems with finite impulse response structure is studied. Expressions for the variance are derived using asymptotic methods. The variance expressions explicitly show the influence of step length, input spectrum, noise variance, parameter variations and model order.