TU Delft education system is transformed on three levels: 1) new courses and projects in existing B.Sc. and M.Sc. programs for multidisciplinary and reflective learning; 2) new M.Sc. programs focusing on multi and interdisciplinarity, personal development, and professional skills; and 3) central Interdisciplinary Projects for Master Students from different programs. With these steps, the university offers students a learning ecosystem where identity-building can occur, fosters interdisciplinary teamwork, and strong interaction with the professional world and government is necessary to finish projects. In this article, the ecosystem will be explained, and results will be shared of surveys among students who experienced learning in the learning ecosystem. The surveys show that students under stand their future role in the community as engineers, feel that they have acquired new skills, feel better about framing complex problems, and are more competent to work in the industry.

University students are asked to become all-round human beings, knowing how to be engaged in Engineering in the future, as well as wholly socialised and going through personal development steps. However, how and where are the students supposed to acquire these skills? Do we already have them in the Higher Education programmes and curricula? This article explores low threshold steps that can be taken to tweak the curriculum and implicit professionalisation of staff towards incorporating transversal skills and reflective activities that allow students to develop to their full potential.. One is a roadmap Workshop identifying guiding principles and touchpoint activities for curricular change. The other is a survey on how transversal skills are currently thought to have been embedded in the curriculum.

Lane detection serves as a fundamental task for automated vehicles and Advanced Driver Assistance Systems. However, current lane detection methods can not deliver the versatility of accurate, robust, and realtime compatible lane detection in real-world scenarios especially under challenging driving scenes. Available vision-based methods in the literature do not consider critical regions of the image and their spatial-temporal salience regarding the detection results, thus they deliver poor performance in peculiar difficult circumstances (e.g., serious occlusion, dazzle lighting). This study aims to introduce a novel sequential neural network model with a spatial-temporal attention mechanism that can focus on key features of lane lines and exploit salient spatial-temporal correlations among continuous image frames for the purpose of enhancing the accuracy and robustness of lane detection. Under the regular encoder-decoder structure and with the implementation using common neural network backbones, the proposed model is trained and evaluated on three large-scale opensource datasets. Extensive experiments demonstrate the strength and the robustness of the proposed model outperforming available state-of-the-art methods in various testing.

In this paper we studied the student’s perception of the acquisition of professional capabilities in Challenge based learning environments with a strong reflective component. The results show students feel the relevance of personnel development from the very moment the enter their master studies. However, they only truly acquire all the relevant professional capabilities when working in interdisciplinary teams on real life problems in interaction with stakeholders.

In this research a controller is developed that can control path-tracking both within and beyond stable limit handling. The controller is based on the equations of motion of the nonlinear bicycle model. The performance of the controller is evaluated in simulation, a sensitivity analysis is performed and the controller is implemented on a 1/10 scale radio controlled car. The controller is able to track a path in normal driving conditions and let the vehicle enter and maintain a drift while remaining close to the desired path.

By installing a battery storage system in the power grid, Distribution Network Operators (DNOs) can solve congestion problems caused by decentralized renewable generation. This paper provides the necessary theory to use such a community battery for grid congestion reduction, backed up by experimental results. A simple network model was constructed by linearizing the load flow equations using a constant impedance load model. Using this model, an accurate estimate of voltage and overload problems is fed into a receding horizon charge path optimizer. The charge path optimization problem is posed as a linear problem and subsequently solved by an LP solver. The algorithms have been applied and validated on a real-world community battery installation. It was found that the voltages and currents can be controlled to a great degree, increasing the grid capacity significantly. The proposed control framework can be used to safeguard network constraints and is compatible with other battery control goals, such as energy trading or energy independence. Network design formulas are described with which a DNO can quickly estimate the potential (de) stabilization of a community battery on the steady-state voltages and currents in the grid.

In this research a controller is developed that can control path-tracking both within and beyond stable limit handling. A controller is developed, based on the equations of motion of the nonlinear bicycle model. The performance of the controller is evaluated in both simulation and on a 1/10 scale radio controlled car. The controller is able to track a path in typical cornering conditions and let the vehicle enter and maintain a drift while remaining close to the desired path.

At Polytechnics design & engineering students are taught about state-of-the-art technical knowledge. Students become qualified engineers and learn to innovate artifacts related to their domain. Not taught is how to develop new engineering knowledge within a multidisciplinary context of stakeholders, companies and regulations. In short, students don't learn to innovate technology. What is taught today is the result of a technological innovation of yesterday. This is not sufficient for industry to innovatively deal with society's grand challenges. The paper describes a project that aims to educate all TU Delft graduate students in the verb of innovating technology, that is, the development of new technologies from inventions in the labs to full- fledged application in business. Such along three dimensions: technical, human and business. The educational portfolio consists of three modules in line with growth along Bloom's taxonomy and online materials on theoretical backbones. All modules apply the notion of technological innovation journeys (Tijo's). Tijo's are rich descriptions of the developmental journey of new technology and are based on inventions from the university's own labs.

Modeling low voltage networks poses a challenge to Distribution System Operators (DSOs) because the low voltage networks generally consists of millions of cables. This paper provides a method to model congestion problems and applies this to such a large low voltage network. By modifying the load model of a customer, a linear load flow model was created. Using a custom sparse solver model, all instantaneous currents and voltages were calculated for the network of Liander DSO, containing over 20 million cables and 3 million power customers. The model took only 30 seconds to simulate the entire network. The results shows that the network of Liander DSO can accommodate quite a large number of solar power installations with relative ease. Also, stepchange transformers are shown to have quite some potential to solve voltage issues that can arise due to solar power.

In this paper we develop a scenario-based Distributed Model Predictive Control (DMPC) approach for large-scale freeway networks. The uncertainties in a large-scale freeway network are categorized into global uncertainties for the overall network and local uncertainties for subnetworks. A reduced scenario tree is proposed, consisting of global scenarios and a reduced local scenario tree. For handling uncertainties in the scenario-based DMPC problem, a min-max setting is considered. A case study is implemented for investigating the scenario-based DMPC approach, and the results show that in the presence of uncertainties it is effective in improving the control performance with the queue length constraint being satisfied.

To deal with the traffic congestion and emissions, traffic-responsive control approaches can be used. The main aim of the control is then to use the existing capacity of the network efficiently, and to reduce the harmful economical and environmental effects of heavy traffic. In this paper, we design a highly efficient model-predictive control system that uses a gradient-based approach to solve the optimization problem, which has been reformulated as a two-point boundary value problem. A gradient-based approach computes the derivatives to find the optimal value. Therefore, the optimization problem should involve only smooth functions. Hence, for nonsmooth functions that may appear in the internal model of the MPC controller, we propose smoothening approaches. The controller then uses an integrated smooth flow and emission model, where the control objective is to reduce a weighted combination of the total time spent and total emissions of the vehicles. We perform simulations to compare the efficiency and the CPU time of the smooth and nonsmooth optimization approaches. The simulation results show that the smooth approach significantly outperforms the nonsmooth one both in the CPU time and in the optimal objective value.

With an increasing use of distributed energy resources and intelligence in the electricity infrastructure, the possibilities for minimizing costs of household energy consumption increase. Technology is moving toward a situation in which households manage their own energy generation and consumption, possibly in cooperation with each other. As a first step, in this paper a decentralized controller based on model predictive control is proposed. For an individual household using a micro combined heat and power (μCHP) plant in combination with heat and electricity storages the controller determines what the actions are that minimize the operational costs of fulfilling residential electricity and heat requirements subject to operational constraints. Simulation studies illustrate the performance of the proposed control scheme, which is substantially more cost effective compared with a control approach that does not include predictions on the system it controls.