CO2 emission of vehicles and its influence on climate change is a widely discussed topic already for many years. New CO2 emission norms for vehicles have been defined based on the 2021 baseline: 15% reduction resulting in 0.09 kg/km for 2025, and 37.5% reduction resulting in 0.069 kg/km for 2030. Both norms are defined under the Worldwide harmonized Light-duty Test Procedure (WLTP) and apply to light-duty vehicles. In this study a new accurate measure is introduced to determine data related CO2 emissions of vehicle automation. The data related aspects in this study include: sensing components, the computing platform, disks inside the vehicle, wireless communication networks and data centers. These data related emissions are used to compare to the norms, which are based on the propulsion of the vehicle. A computational model is developed based on literature and a survey to obtain the instantaneous CO2 emission rates (kg CO2-e/h) of the data related aspects for varying scenarios. Scenario analysis is applied to define six different scenarios. The scenarios differ based on two varying sensing compositions and three varying energy grids. The two sensing compositions correspond to a Tesla Model S (SAE level 2) and Waymo's Chrysler Pacifica (SAE level 4). The different energy grids correspond to the 2019 Climate Act targets for 2030 and 2050 to reduce CO2 emission, and are referred to as fixed CO2 emission rates in kg CO2-e/Wh. Sensitivity analysis of the instantaneous CO2 emission rate is applied, to obtain the data related aspects that have the strongest influence on the resulting CO2 emission in each scenario. The instantaneous data related CO2 emission rates are translated to the driving cycle of the WLTP to obtain the spatial data related CO2 emission rates (kg CO2-e/km), which are used to compare to the defined propulsion-based CO2 norms of vehicles. From the sensitivity analysis, it is concluded that the energy intensity of wireless communication networks and the data transmission rate from vehicle to data center, are the two data related aspects that have the strongest influence on the instantaneous CO2 emission rate. From the spatial CO2 emission rate of each scenario, it is concluded that the energy grid also significantly affects whether the norms can be met. For high amounts of data transmission, the norms seem to be difficult to achieve in most scenarios and situations. Therefore, the energy consumption of wireless communication networks should be further optimized. Regarding the data transmission rate, further research should be obtained to gain more insights into the values of the data transmission rates dependent on the travel speed. Based on the assumptions made in this study, the CO2 emission from the data related aspects seem to exceed the propulsion-based CO2 norms of vehicles. Future work should investigate if data related aspects of vehicle automation need to be included in the CO2 norms of vehicles.

Optimization of traffic signal control has been widely investigated by means of model-based strategies. In 2012 a new model-based controller was published, named Schedule-driven Intersection Control (SCHIC). This controller uses a job-scheduling algorithm to minimize the cumulative delay for all observed vehicles. The algorithms of SCHIC are at the basis of the commercially deployed controller SURTRAC. This means that SCHIC is partly used in real-world traffic situations and its algorithms are well documented. Therefore, it was possible to reproduce this controller in this thesis. Previous literature often evaluated SCHIC in simplified scenarios with (1) unrealistically large look-ahead times for vehicle detection. Secondly, (2) in most scenarios all approaching vehicles were restricted to request a single turning-movement at the intersection thereby removing inter-phase uncertainty (as explained directly below). Inter-phase uncertainty occurs when the chosen entry-lane of a vehicle determines the green phase that it requires from the traffic light controller. So, inter-phase uncertainty occurs under the following two conditions. Condition 1 – an approaching vehicle must eventually choose between different entry-lanes, with each entry-lane being associated to a different turning-movement at the intersection. Condition 2 – each of these individual entry-lanes is serviced during a different and competing green phase. For example, inter-phase uncertainty occurs when two competing green phases service either the straight-ahead or the left-turning entry-lane. The aim of this thesis is to evaluate SCHIC with both (1) realistic look-ahead times and (2) inter-phase uncertainty, since these two uncertainties are often not found in previous literature, however, they are commonly found in reality. Thus, it is not directly clear how the reported results would translate to more realistic settings, which include these two common uncertainties. In this study SCHIC is re-implemented and evaluated on a typical Dutch urban intersection located in The Hague. The lane dimensions of this intersection determine the look-ahead times as used in this thesis. For this study, a new method is introduced to reduce inter-phase uncertainty which is inspired by lane change prediction modules for automated driving. This method, Lane Change Prediction (LCP), makes use of e.g. radar-equipment in order to observe: (1) the driving style of each approaching vehicle and (2) the upcoming traffic state on each reachable entry-lane. Based on these two observations a classification model, i.e. a Random Forest, predicts the desired entry-lane before the vehicle actually enters it. This thesis researches whether the potentially higher accuracy of LCP will improve SCHICs performance as compared to its simpler statistical baselines, fixed-label and split-vehicles, which could have a lower accuracy. All traffic scenarios are simulated with Simulation of Urban Mobility (SUMO). Two popular driver-models are evaluated which determine the dynamic behaviour of the vehicles in simulation, i.e. the Intelligent Driver Model (IDM) and Krauss driver-model. The results showed that despite its higher accuracy, LCP does not significantly improve SCHICs performance over its baseline methods fixed-label and split-vehicles. Furthermore, the maximum theoretical performance gain of using LCP is evaluated with perfect predictions (Oracle). When applying perfect predictions, the average speed of all vehicles increased by 0.65% and 0.51% for the Intelligent Driver Model (IDM) and Krauss driver-model, respectively. The average number of stops per vehicle was reduced by 1.53% and 1.12% for the IDM and Krauss driver-model, respectively. Overall, it is concluded that the performance potential of all methods, i.e. the {Oracle, split-vehicles, fixed-label and LCP}, is quite comparable on SCHICs traffic flow. These methods show some performance potential, however, their confidence intervals are relatively wide which draws them closer to each other. In our scenario, the intersection in The Hague, another method might increase SCHICs traffic flow more significantly.

Lane change decision-making is an important challenge for automated vehicles, urging the need for high performance algorithms that are able to handle complex traffic situations. Deep reinforcement learning (DRL), a machine learning method based on artificial neural networks, has recently become a popular choice for modelling the lane change decision-making process, outperforming various traditional rule-based models. So far, performance has often been expressed in terms of achieved average speed, absence of collisions or merging success rate. However, no studies have investigated how humans will react to the resulting behavior as potential occupants. This study addresses this research gap by validating a self-developed DRL-based lane changing model (trained using proximal policy optimization) from a technology acceptance perspective through an online crowdsourcing experiment. Participants (N=1085) viewed a random subset of 32 out of 120 videos of an automated vehicle driving on a three-lane highway with varying traffic densities featuring our proposed model or a baseline policy (i.e. a state-of-the-art rule-based model, MOBIL). They were tasked to press a response key if the decision-making was deemed undesirable and subsequently rated the vehicle's behavior along four acceptance constructs (performance expectancy, safety, human-likeness and reliability) on a scale of 1 to 5. Results showed that the proposed model caused a significantly lower amount of disagreements and was rated significantly higher on all four acceptance constructs compared to the baseline policy. Moreover, considerable differences between individual disagreement rates were observed for both models. Our findings offer prospects for the practical application of DRL-based lane change models in a use-case scenario, depending on the user. Further research is necessary to examine whether these observations hold in other (more complex) traffic situations. Additionally, we recommend combining DRL with other modelling techniques that allow for personalization of behavioral parameters, such as imitation learning.

Truck platooning refers to coordinating a group of heavy-duty vehicles at a close inter-vehicle distance to reduce overall fuel consumption. This coordination between trucks is traditionally achieved by adjusting the schedule, velocity, and routines to increase the platooning chances, and thus improve the overall fuel efficiency. However, the data model built for the coordination problem is typically integer-constrained, making it generally hard to solve. On the other hand, the interaction among self-interested fleets that are operated by different companies is not well-studied. This thesis aims to build a distributed framework for multi-fleet truck platooning coordination to enable the coordination without a third-party service provider. The interaction among fleets is considered a non-cooperative finite game, for which we propose the best response search method, which essentially requires to solve a cooperative truck platooning optimization problem iteratively. We refer to the optimization problem as a best-response problem, which is formulated as a mixed-integer linear problem with relaxation skills. To achieve a feasible time complexity for the best-response subproblem, we propose a decentralized algorithm, distributing the computational load to connected automated vehicles within the fleet. The proposed method is examined under a real-world featured demand set to compare the performance in optimality and time complexity with previous studies. The result suggests that the decentralized algorithm delivers the optimal objective value in this case, while the best-response search does not deliver extra benefits as the dominating time costs in the cost functions eliminate the potential for improvement.

Growth of mobility for people and goods transportation has been increasing steadily over the years. With the perpetual increase in number of road vehicles, comes the inevitable problems of traffic and pollution. Various intelligent traffic technologies and strategies have been proposed and implemented over the years to overcome the road traffic problems.  Platooning is identified as one of the ways to tackle transportation related problems efficiently. Even though platooning has been tested and implemented in highways, not much attention has been given to controlling the platoons at urban roads. When vehicles in the platoon are connected and automated, it helps them to understand the environment better and communicate efficiently for better performance. Platoons at the vicinity of signalized intersections need to accelerate and decelerate in a nonlinear manner which leads to higher energy consumption and longer travel time. Most of the previous research approaches focused on optimizing only the energy consumption or travel time of the platoons. However, studies show that the recommended driving advice from the controllers is not tracked by the drivers completely. The current research focuses on optimal trajectory planning of the electric vehicle platoons at signalized intersections. This is done by designing an energy efficient longitudinal controller for the platoons using optimal control method. The goal of this thesis is split between planning and tracking the trajectory. Thus the optimal speed profile is planned by the high level controller (i.e the optimal controller) and tracked using a battery electric vehicle model controlled by a low level controller (i.e a PID controller). The performance of the controlled platoon was verified using different scenarios and was found to perform positively under respective control objectives and constraints. The designed controlled and automated platoon by the optimal controller was able to achieve energy consumption and cost saving up to 67% when compared with intelligent driver model (IDM) platoon for a specific scenario. This research is a collaboration between the Research Institute of Highways (RIOH) in China and Rijkswaterstaat, Ministry of Infrastructure and Water Management in The Netherlands, to overcome the common traffic problems on road. The research was carried out at the ITS-Edulab, a Dutch traffic and transport laboratory for students. It is a cooperation between Rijkswaterstaat and the Transport & Planning group of the Delft University of Technology.

Traffic congestion on highways is a multi-sectoral phenomenon affecting society, the economy and the environment. It often takes place at specific locations such as on and off-ramps, weaving segments and intersections. The on-ramp merging procedure is considered as one of the main factors that causes traffic congestion on highways. The studies in literature show that the merging procedure can result in adverse traffic scenarios such as the buildup of the vehicles on the ramp which causes a downstream drop in capacity and subsequent blockage of upstream off-ramp traffic flow. Moreover, the on-ramp vehicles need to take the actions of leading and following mainlane vehicles into account during the merging process. On highly congested roads, this merging process becomes even more tedious and undesirable stop-and-go traffic behavior becomes unavoidable. Connected and autonomous vehicles (CAVs) that can provide safe gaps between vehicles along with identifying appropriate merging speed profiles have the potential to reduce traffic accidents and improve traffic efficiency. This thesis introduces a nonlinear model predictive control (NMPC) strategy for autonomous merging control based on a cost function that tracks the desired inter-vehicular gaps for on-ramp and mainlane vehicles, and thus intends to fully exploit the capacity of the road in order to maximize the traffic throughput. The proposed controller aims to optimize both acceleration and steering rate profiles of vehicles, and to guide on-ramp vehicles to merge efficiently, without frequent slowdown or wait for merging gaps at the end of the ramp along with minimal disruption to the mainlane traffic flow. The controller is evaluated under different initial conditions, ranging from low to high traffic conditions. The performance of the controller is compared to that of a baseline scenario, and the results show that the proposed controller increases travel times in the range of 2.46% and 4.17% for different traffic conditions, without disrupting the mainline traffic operation. Additionally, average speed of vehicles is improved in the range of 8.2% and 4.5% under different traffic conditions.

Traffic congestion causes detrimental effects on economy and society in terms of travel time delay, increased vehicle collision risk and increased air pollution. To tackle that, new intelligent transportation systems are being developed. One of these emerging technologies comprise automated driving systems. Adaptive Cruise Control (ACC), enabling a vehicle to follow its leader automatically, is an automated driving system which has been available in the vehicle market. Literature shows that traffic flow performance may not significantly benefit from ACC due to traffic instability and large headways. As an extension of ACC, Cooperative ACC is designed to improve traffic stability and throughput by using Vehicle-to-Vehicle (V2V) communication. Thanks to the anticipation of the downstream traffic, a short following gap can be realized by CACC which is highly expected to considerably increase road capacities. Before CACC vehicle is allowed and promoted in the market, it is crucial for policy makers and road operators to gain insights into the traffic flow impacts of CACC systems. Existing studies show that CACC can increase traffic throughput at high vehicle market penetration. However, the realistic effects of CACC on traffic flow have not been adequately revealed because the behaviour of CACC vehicles has not been realistically modelled. The multiple CACC driving modes, the degraded operation to ACC and the human driver take-over control when it is out of the CACC operational design domain, have not yet been explicitly included in existing CACC impact studies. A scientific gap prevails in modelling the complex CACC behaviour and investigating its actual influence on traffic flow especially at realistic networks with a single bottleneck and interacting bottlenecks.

The Dutch governmental organisation Rijkswaterstaat contributes to the smooth and safe flow of traffic, as both traffic jams and accidents cost society large amounts of money each day. Roads are designed for the current traffic composition. Due to the promotion of Adaptive Cruise Control (ACC) systems, utilisation of these systems is expected to increase. Society benefits from insights into the effects these systems have on traffic flow, as they can help to reduce traffic jams and accidents. ACC systems are designed to increase driving comfort by taking over throttling and braking from the human driver. For optimal driver acceptance, these systems show similar driving behaviour to that of human drivers. However, this is not entirely possible due to limited anticipation. To predict how differences in driving behaviour affect traffic flows, researchers usually perform simulations using parametric car-following models. However, research shows contradictory findings. The goal of this research was to gain insights into the performance of commonly applied parametric car-following models on representing the driving behaviour of ACC systems. Optimal model calibration was obtained by investigating the sensitivity of the model calibration to synthetic data. Investigated were the calibration methodology and the quality and quantity of calibration data. Models are calibrated to real-world driving data from an Audi A4 from 2017. These models were used to assess the capability of representing typical highway scenarios: steady-state car-following, cut-in, cut-out, hard-braking and stop-and-go scenarios. The considered models were the Intelligent Driver Model (IDM) model, which has previously been applied to model the driving behaviour of human drivers, the newly developed simplified ACC (sACC) model and a variant on this model. Insights in the sensitivity of the model calibration were obtained by performing a sensitivity analysis on synthetic data. Essential factors in achieving an optimal model calibration are: 1) the model closely matches the driving behaviour in the data, 2) noise levels are as low as possible and 3) the data should contain as many situations as possible that are also included in the model. The dataset must be sufficiently long to include all these situations and to allow the model to develop its dynamics entirely. Using these insights, a calibration was performed on real-world ACC driving data from an Audi A4 (2017). For the ACC system, it was found: 1) the ACC system exhibits non-linear driving behaviour, 2) the acceleration depends on the current velocity and distance to the desired velocity, 3) the system does not consider an intelligent braking strategy and is thus not able of handling safety-critical driving situations and 4) the model includes a sub-controller which ensures comfortable driving behaviour. Except for the comfortable sub-controller, the non-linear IDM model considers all of these factors and thus best represents the driving behaviour. The linear sACC model cannot represent standing conditions, which is resolved in the alternative version. The linearity allows for a better representation of the behaviour of the comfortable sub-controller. However, it disallows for an accurate representation of the dynamics by the models.

The future brings Connected and Automated Vehicles (CAVs), where platooning with small inter-vehicle time gaps is possible. However, before the situation that the entire car park exists of CAVs, there is a transition period. This MSc. thesis has developed two strategies (in MOTUS), to let CAVs actively form platoons with dynamic dedicated lane sections (DDLSs). This means that the lane sections temporarily become dedicated to CAVs only under certain traffic density conditions.

Truck platooning is seen as a method to reduce truck emissions and labor hours in the transport industry. The effect that truck platooning will have on the other traffic receives less attention. In this thesis a methodology is proposed to model truck platoons, including a platoon level controller for lane changing. The model is used to estimate the on-ramp length that vehicles require in order to safely merge into the highway while interacting with platoons and the model is used to estimate the on-ramp length that truck platoons require in order to merge onto the highway if they would do so in platoon formation. Results show that onramps need to be at least 300 meters for other traffic to safely into the highway. For platoons to safely merge into the highway the required distance is heavily dependent on the crowdedness of the highway, ranging between 400-1000 meters of required on-ramp length.

The autonomous transit market is on the rise. 2getthere is one of the lead suppliers of automated people moving systems, located in the Netherlands. Their vehicles can be classified as SAE level 4 automation (Highly Automated) vehicles. This thesis is concerned with the dynamic trajectory generation for 2getthere’s Group Rapid Transport (GRT) mini-bus in case deviation from a nominal trajectory is required.

Trajectory generation can be achieved using several planning approaches. However, only few of these approaches can take into account the vehicle’s dynamics, as well as comfort and safety constraints. Moreover a degree of optimality of the deviation is required in order to keep the vehicle as close as possible to the nominal trajectory. A model predictive control (MPC) based trajectory planner is designed to solve a constrained optimal control problem. Using the ACADO toolkit, highly efficient self contained C code is generated to solve the trajectory generation problem very rapidly.

A single-track model with front and rear wheel steering is used to predict the four-wheel-steered vehicle’s manoeuvring capabilities in every consecutive initial state. This enables the planner to take into account the GRT’s dynamics, without overcomplicating the optimization problem. The performance of the planner is tested in Simulink.

In this research, insight is gained into the impacts of truck platooning on traffic performance and safety at motorway on-ramps in mixed traffic using microscopic simulation. Thereby different platooning strategies and platoon configurations are considered, that for instance differ in desired platoon inter-vehicle gaps and platoon sizes. The effects are considered for a broad range of traffic intensities and penetration rates of equipped trucks. Merging problems are observed, but these can be solved if the platoon members yield to create a gap for merging vehicles.

This study explores the possibility of developing a short-term traffic flow prediction model that can be used to convert installed adaptive controllers to predictive controllers with minimal hardware changes. By using the prediction model’s outputs to virtually trigger vehicle loop detectors, the outputs of an adaptive controller can be extracted in advance of actual vehicle arrivals. This will enable service providers to send out time to green/red (T2G/R) information or green light optimal speed advice (GLOSA), which are driver assistance use cases that aim to efficiently guide vehicles through intersections, in anticipation of known upcoming signal states. The main requirements for developing the prediction model are that it should be scalable to different intersection configurations, adaptable to different traffic conditions and should encompass the nonlinearity of traffic flow behavior. Since it fits these criteria, the developed model is a multi-task learning recurrent neural network with exogenous inputs (NARX), which is designed to match the traffic flow simulation abilities of a well adopted analytical traffic flow progression model. Both models were tested for a corridor in Delft that experiences almost consistent free flow conditions, and on an intersection in Haarlem with varying traffic conditions. For both case studies the neural network outperformed the analytical model. Most notably, it was better adaptable to long queues at intersections, had lower average error values, made fewer large errors, and better recognized the effect of a source/sink. When interfaced with an adaptive controller to test the predictive control methodology proposed, the superiority of the designed neural network model over the analytical model became more prominent. At no point did the neural network’s prediction errors result in queue spill-back, which was not true for the analytical model. However, the overall accuracy of the NARX model was still not yet satisfactory enough for practical application (especially for highly under-saturated traffic conditions) without the use of corrective measures. Nonetheless, due to its significant superiority over the widely used analytical model, with regards to both accuracy and adaptability, this model can be considered as a new starting point for traffic flow progression modeling.