Autonomous driving is an emerging area that has been receiving an increasing amount of interest from different companies and researchers. 3D point cloud map is a significant foundation of autonomous driving as it provides essential information for localization and environment perception. However, when trying to gather road information for map creation, the presence of dynamic objects like vehicles, pedestrians, and cyclists will add noise and unnecessary information to the final map. In order to solve the problem, this thesis presents a novel two-stage model that contains a scan-to-scan removal stage and a scan-to-map generation stage. By designing the new three-branch neural network and new attention-based fusion block, the scan-to-scan part achieves a higher mean Intersection-over-Union (mIoU) score. By improving the ground plane estimation, the scan-to-map part can preserve more static points while removing a large number of dynamic points. The test on SemanticKITTI dataset and Scania dataset shows our two-stage model outperforms other baselines. / Autonom körning är ett nytt område som har fått ett allt större intresse från olika företag och forskare. Kartor med 3D-punktmoln är en viktig grund för autonom körning eftersom de ger viktig information för lokalisering och miljöuppfattning. När man försöker samla in väginformation för kartframställning kommer dock närvaron av dynamiska objekt som fordon, fotgängare och cyklister att lägga till brus och onödig information till den slutliga kartan. För att lösa problemet presenteras i den här avhandlingen en ny tvåstegsmodell som innehåller ett steg för borttagning av skanningar och ett steg för generering av skanningar och kartor. Genom att utforma det nya neurala nätverket med tre grenar och det nya uppmärksamhetsbaserade fusionsblocket uppnår scan-to-scan-delen högre mean Intersection-over-Union (mIoU)-poäng. Genom att förbättra uppskattningen av markplanet kan skanning-till-kartor-delen bevara fler statiska punkter samtidigt som ett stort antal dynamiska punkter avlägsnas. Testet av SemanticKITTI-dataset och Scania-dataset visar att vår tvåstegsmodell överträffar andra baslinjer.

A requirement for safe autonomous driving is to have an accurate global localization of the ego vehicle. Methods based on Global Navigation Satellite System (GNSS) are the most common but are not precise enough in areas without good satellite signals. Instead, methods likelandmark-based localization (LBL) can be used. In LBL, sensors onboard the vehicle detectlandmarks near the vehicle. With these detections, the vehicle’s position is deduced by looking up matching landmarks on a high-definition map. Commonly found along roads, stretching for long distances, guardrails are a great landmark that can be used for LBL. In this thesis, two different methods are proposed to detect and vectorize guardrails from vehicle sensor data to enable future map matching for LBL. The first method uses semantically labeled LiDAR data with pre-classified guardrail LiDAR points as input data. The method is based on the DBSCAN clustering algorithm to cluster and filter out false positives from the pre-classified LiDAR points. The second algorithm uses raw LiDAR data as input. The algorithm finds guardrail candidate points by segmenting high-densityareas and matching these with thresholds taken from the geometry of guardrails. Similar to the first method, these are then clustered into guardrail clusters. The clusters are then vectorized into the wanted output of a 2D vector, corresponding to points inside the guardrail with aspecific interval. To evaluate the performance of the proposed algorithms, simulations from real-life data are analyzed in both a quantitative and qualitative way. The qualitative experiments showcase that both methods perform well even in difficult scenarios. Timings of the simulations show that both methods are fast enough to be applicable in real-time use cases. The defined performance measures show that the method using raw LiDAR data is more robust and manages to detect more and longer parts of the guardrails.

In order to reach higher levels of autonomy in autonomous driving, it is important to consider potential occluded traffic participants. Current research has considered occlusion-aware autonomous driving in urban situations. However, no implementations have shown good performance in high velocity situations such as highway driving yet, since the current methods are too conservative in these situations and result in frequent excessive braking. In this work a method is proposed that tracks boundaries on the velocity states of potential hidden traffic using reachability analysis. It is proven that the method can guarantee collision-free trajectories for any, potentially hidden, traffic. The method is evaluated on cut-in scenarios retrieved from a dataset of recorded traffic. The results show that tracking the velocity bounds for potentially hidden traffic results in more efficient trajectories up to 18 km/h faster compared to existing occlusion-aware methods. While the method shows clear improvements, it does not always manage to establish a velocity bound and at times excessive braking still occurs. Further work is thus necessary to ensure consistently well-performing occlusion-aware highway driving. / För att nå högre nivåer av autonomi vid autonom körning är det viktigt att ta hänsyn till möjliga skymda trafikanter. Aktuell forskning har övervägt skymd-sikt-medveten autonom körning i urbana situationer. Emellertid har inga implementeringar visat bra prestanda i höghastighetssituationer såsom motorvägskörning ännu, eftersom de nuvarande metoderna är för konservativa i dessa situationer och resulterar i frekventa överdrivna inbromsningar. I detta arbete föreslås en metod som bestämmer gränser för hastighetstillstånden för möjlig skymd trafik med hjälp av nåbarhetsanalys. Det är bevisat att metoden kan garantera kollisionsfria banor för all möjlig skymd trafik. Metoden utvärderas på scenarier hämtade från ett dataset av registrerad trafik. Resultaten visar att bestämning av hastighetsgränserna för möjlig skymd trafik resulterar i effektivare banor upp till 18 km/h snabbare jämfört med befintliga skymd-sikt-medvetna-metoder. Även om metoden visar tydliga förbättringar, lyckas den inte alltid fastställa en hastighetsgräns och ibland förekommer fortfarande överdriven inbromsning. Ytterligare arbete är därför nödvändigt för att säkerställa konsekvent välpresterande motorvägskörning under skymd sikt.

<p>In traffic scene perception for autonomous vehicles, driving videos are available from in-car sensors such as camera and LiDAR for road detection and collision avoidance. There are some existing challenges in computer vision tasks for video processing, including object detection and tracking, semantic segmentation, etc. First, due to that consecutive video frames have a large data redundancy, traditional spatial-to-temporal approach inherently demands huge computational resource. Second, in many real-time scenarios, targets move continuously in the view as data streamed in. To achieve prompt response with minimum latency, an online model to process the streaming data in shift-mode is necessary. Third, in addition to shape-based recognition in spatial space, motion detection also replies on the inherent temporal continuity in videos. While current works either lack long-term memory for reference or consume a huge amount of computation. </p> <p><br></p> <p>The purpose of this work is to achieve strongly temporal-associated sensing results in real-time with minimum memory, which is continually embedded to a pragmatic framework for speed and path planning. It takes a temporal-to-spatial approach to cope with fast moving vehicles in autonomous navigation. It utilizes compact road profiles (RP) and motion profiles (MP) to identify path regions and dynamic objects, which drastically reduces video data to a lower dimension and increases sensing rate. Specifically, we sample one-pixel line at each video frame, the temporal congregation of lines from consecutive frames forms a road profile image; while motion profile consists of the average lines by sampling one-belt pixels at each frame. By applying the dense temporal resolution to compensate the sparse spatial resolution, this method reduces 3D streaming data into 2D image layout. Based on RP and MP under various weather conditions, there have three main tasks being conducted to contribute the knowledge domain in perception and planning for autonomous driving. </p> <p><br></p> <p>The first application is semantic segmentation of temporal-to-spatial streaming scenes, including recognition of road and roadside, driving events, objects in static or motion. Since the main vision sensing tasks for autonomous driving are identifying road area to follow and locating traffic to avoid collision, this work tackles this problem by using semantic segmentation upon road and motion profiles. Though one-pixel line may not contain sufficient spatial information of road and objects, the consecutive collection of lines as a temporal-spatial image provides intrinsic spatial layout because of the continuous observation and smooth vehicle motion. Moreover, by capturing the trajectory of pedestrians upon their moving legs in motion profile, we can robustly distinguish pedestrian in motion against smooth background. The experimental results of streaming data collected from various sensors including camera and LiDAR demonstrate that, in the reduced temporal-to-spatial space, an effective recognition of driving scene can be learned through Semantic Segmentation.</p> <p><br></p> <p>The second contribution of this work is that it accommodates standard semantic segmentation to sequential semantic segmentation network (SE3), which is implemented as a new benchmark for image and video segmentation. As most state-of-the-art methods are greedy for accuracy by designing complex structures at expense of memory use, which makes trained models heavily depend on GPUs and thus not applicable to real-time inference. Without accuracy loss, this work enables image segmentation at the minimum memory. Specifically, instead of predicting for image patch, SE3 generates output along with line scanning. By pinpointing the memory associated with the input line at each neural layer in the network, it preserves the same receptive field as patch size but saved the computation in the overlapped regions during network shifting. Generally, SE3 applies to most of the current backbone models in image segmentation, and furthers the inference by fusing temporal information without increasing computation complexity for video semantic segmentation. Thus, it achieves 3D association over long-range while under the computation of 2D setting. This will facilitate inference of semantic segmentation on light-weighted devices.</p> <p><br></p> <p>The third application is speed and path planning based on the sensing results from naturalistic driving videos. To avoid collision in a close range and navigate a vehicle in middle and far ranges, several RP/MPs are scanned continuously from different depths for vehicle path planning. The semantic segmentation of RP/MP is further extended to multi-depths for path and speed planning according to the sensed headway and lane position. We conduct experiments on profiles of different sensing depths and build up a smoothly planning framework according to their them. We also build an initial dataset of road and motion profiles with semantic labels from long HD driving videos. The dataset is published as additional contribution to the future work in computer vision and autonomous driving. </p>

Autonomous driving is one of the key technologies for increasing road safetyand reducing traffic volumes. Therefore, science and industry are workingtogether on new innovative solutions in this field of technology. One importantcomponent in this context is the approval and testing of new solution concepts,with special focus on the ones for urban environments. Not only because ofthe high diversity of traffic situations, but also because of the close contactbetween vulnerable road users (VRU) and automated vehicles.In the course of this work, a novel approach for testing automated drivingfunctions and vehicle systems in urban environments is presented. The goal isto create a safe and valid environment in which the automated vehicle and theVRU can meet and interact. The basis is a highly realistic virtual model of acity center. The physical behavior of the vehicle and VRU is recorded usingmeasurement technology and transferred to the virtual city model.Based on representative urban traffic scenarios, the functionality of the urbantest field is investigated from various points of view. Thereby, the focus is onreal-time capability and the quality of interaction between the vehicle and theVRU.The investigations show that both the real-time capability and the interactionpossibilities could be demonstrated. Further, the developed methodologies aresuitable for real time applications. / CityInMotion

This work proposes a strategy for visual perception in the context of autonomous driving. Despite the growing research aiming to implement self-driving cars, no artificial system can claim to have reached the driving performance of a human, yet. Humans---when not distracted or drunk---are still the best drivers you can currently find. Hence, the theories about the human mind and its neural organization could reveal precious insights on how to design a better autonomous driving agent. This dissertation focuses specifically on the perceptual aspect of driving, and it takes inspiration from four key theories on how the human brain achieves the cognitive capabilities required by the activity of driving. The first idea lies at the foundation of current cognitive science, and it argues that thinking nearly always involves some sort of mental simulation, which takes the form of imagery when dealing with visual perception. The second theory explains how the perceptual simulation takes place in neural circuits called convergence-divergence zones, which expand and compress information to extract abstract concepts from visual experience and code them into compact representations. The third theory highlights that perception---when specialized for a complex task as driving---is refined by experience in a process called perceptual learning. The fourth theory, namely the free-energy principle of predictive brains, corroborates the role of visual imagination as a fundamental mechanism of inference. In order to implement these theoretical principles, it is necessary to identify the most appropriate computational tools currently available. Within the consolidated and successful field of deep learning, I select the artificial architectures and strategies that manifest a sounding resemblance with their cognitive counterparts. Specifically, convolutional autoencoders have a strong correspondence with the architecture of convergence-divergence zones and the process of perceptual abstraction. The free-energy principle of predictive brains is related to variational Bayesian inference and the use of recurrent neural networks. In fact, this principle can be translated into a training procedure that learns abstract representations predisposed to predicting how the current road scenario will change in the future. The main contribution of this dissertation is a method to learn conceptual representations of the driving scenario from visual information. This approach forces a semantic internal organization, in the sense that distinct parts of the representation are explicitly associated to specific concepts useful in the context of driving. Specifically, the model uses as few as 16 neurons for each of the two basic concepts here considered: vehicles and lanes. At the same time, the approach biases the internal representations towards the ability to predict the dynamics of objects in the scene. This property of temporal coherence allows the representations to be exploited to predict plausible future scenarios and to perform a simplified form of mental imagery. In addition, this work includes a proposal to tackle the problem of opaqueness affecting deep neural networks. I present a method that aims to mitigate this issue, in the context of longitudinal control for automated vehicles. A further contribution of this dissertation experiments with higher-level spaces of prediction, such as occupancy grids, which could conciliate between the direct application to motor controls and the biological plausibility.

Traffic coordination is an essential challenge in vehicle automation. The challenge is not only about maximizing the revenue/productivity of a fleet of vehicles, but also about avoiding non feasible states such as collisions and low energy levels, which could make the fleet inoperable. The challenge is hard due to the complex nature of the real time traffic and the large state space involved. Reinforcement learning and simulation-based search techniques have been successful in handling complex problem with large state spaces [1] and can be used as potential candidates for traffic coordination. In this degree project, a variant of these techniques known as Dyna-2 [2] is investigated for traffic coordination. A long term memory of past experiences is approximated by a neural network and is used to guide a Temporal Difference (TD) search. Various features are proposed, evaluated and finally a feature representation is chosen to build the neural network model. The Dyna-2 Traffic Coordinator (TC) is investigated for its ability to provide supervision for handling vehicle bunching and charging. Two variants of traffic coordinators, one based on simple rules and another based on TD search are the existing baselines for the performance evaluation. The results indicate that by incorporating learning via a long-term memory, the Dyna-2 TC is robust to vehicle bunching and ensures a good balance in charge levels over time. The performance of the Dyna-2 TC depends on the choice of features used to build the function approximator, a bad feature choice does not provide good generalization and hence results in bad performance. On the other hand, the previous approaches based on rule-based planning and TD search made poor decisions resulting in collisions and low energy states. The search based approach is comparatively better than the rule-based approach, however it is not able to find an optimal solution due to the depth limitations. With the guidance from a long term memory, the search was able to generate a higher return and ensure a good balance in charge levels. / Trafikkoordinering är en grundläggande utmaning för att autonomisera fordon. Utmaningen ligger inte bara i att maximera inkomsten/produktiviteten hos en fordonsflotta utan även i att undvika olämpliga tillstånd, så som krockar och brist på energi vilka skulle kunna göra flottan obrukbar. Utmaningen är svår på grund av den komplexa naturen hos trafik i realtid och det stora tillståndsrummet som innefattas. Förstärkningsinlärning och simulationsbaserade söktekniker har varit framgångsrika metoder för att hantera komplexa problem med stora tillståndsrum [1] och kan ses som en potentiell kandidat för trafikkoordinering. Detta examensarbete undersöker en variant av dessa tekniker, känd som Dyna-2 [2], applicerat på trafikkoordinering. Ett långsiktigt minne av tidigare erfarenheter approximeras med ett neuron nät och används för att vägleda en Temporal Difference (TD) sökning. Olika attribut föreslås, utvärderas och sätts sedan samman till en representation att bygga nätverket kring. Dyna-2 Trafikkoordinator (TC) undersöks för dess färdighet att ge beslutsstöd för hantering av grupperade fordon och laddning. Två varianter av trafikkoordinerare, en baserad på enkla regler och en baserad på TD-sökningen, används som grund för utvärderingen av prestanda. Resultaten indikerar att genom inkludering av inlärning via ett långsiktigt minne så är Dyna-2 TC en robust metod för att hantera grupperade fordon och ger en god balans av laddningsnivå över tid. Prestandan hos Dyna-2 TC beror på valet av de attribut som används för att bygga approximeringsfunktionen, sämre val av attribut generaliserar inte bra vilket då resulterar i dålig prestanda. Å andra sidan, de tidigare tillvägagånssätten baserade på planering genom regler och TD-sökning tog dåliga beslut vilket resulterade i kollisioner och tillstånd med låga laddningsnivåer. Jämfört med att basera på regler så är den sökbaserade metoden bättre, den lyckades dock inte hitta en optimal lösning på grund av begränsningar hos sökdjupet. Med vägvisning från ett långsiktigt minne så sökningen kunde sökningen generera högre avkastning och säkerställa en god balans hos laddningsnivåerna.

One central challenge for Autonomous Driving (AD) systems is ensuring functional safety. This is affected by all parts of vehicle automation systems: environment perception, decision making, and actuation. The AD system manages its activity towards achieving its goals to maintain in the safety domain, upon an environment using observation through sensors and consequent actuators. Therefore, this research investigates the operational safety for the AD system. In this research, a simulation for the Autonomous Emergency Braking (AEB) system and a simple scenario are constructed on CARLA, an open-source simulator for autonomous driving systems, to investigate the factors that impact the performance of the AEB system. The time-series data that influence the AEB are collected and fed into three time-series analysis algorithms, Autoregressive Integrated Moving Average model (ARIMA), regression tree and Long short-term memory (LSTM), to select a suitable time-series algorithm to be used for the AEB system. The results show that weather, the measurement range of the sensors, and noise can affect the results of the AEB system. After comparing the performance of these three time-series algorithms through contrasting the recall and precision of these three algorithms to detect noise in the data, the results can be obtained that LSTM has the better performance for long-term analysis. And ARIMA is more suitable for short-term time-series analysis. LSTM is chosen to analyze the time-series data, since the long-term time-series analysis is necessary for the AEB system and it can detect the noise in the variables of the AEB system with better performance. / En central utmaning för AD system är att säkerställa funktionell säkerhet. Detta påverkas av alla delar av fordonsautomatiseringssystem: miljöuppfattning, beslutsfattande och aktivering. AD -systemet hanterar sin aktivitet för att uppnå sina mål att upprätthålla inom säkerhetsområdet, i en miljö som använder observation genom sensorer och därav följande ställdon. Därför undersöker denna forskning den operativa säkerheten för AD systemet. I denna forskning konstrueras en simulering för AEB -systemet och ett enkelt scenario på CARLA, en simulator med öppen källkod för autonoma körsystem, för att undersöka de faktorer som påverkar prestandan för AEB systemet. Tidsseriedata som påverkar AEB samlas in och matas in i tre tidsserieanalysalgoritmer, ARIMA, regressionsträd och LSTM, för att välja en lämplig tidsserie-algoritm som ska används för AEB systemet. Resultaten visar att väder, mätområdet för sensorerna och brus kan påverka resultaten av AEB systemet. Efter att ha jämfört prestandan för dessa tre tidsserie-algoritmer genom att kontrastera återkallelsen och precisionen för dessa tre algoritmer för att detektera brus i data kan resultaten erhållas att LSTM har bättre prestanda för långsiktig analys. Och ARIMA är mer lämpad för korttidsanalyser i tidsserier. LSTM väljs för att analysera tidsseriedata, eftersom långsiktig tidsserieanalys är nödvändig för AEB systemet och det kan detektera bruset i variablerna i AEB system med bättre prestanda.

Mit der Einführung hochautomatisierter Assistenzfunktionen soll Fahrzeugführern in naher Zukunft eine Abwendung von der Fahraufgabe ermöglicht werden. Neben der Steigerung des individuellen Komforts besteht die Erwartung an eine gleichzeitig erhöhte oder zumindest vergleichbare Sicherheitsbilanz im weiterhin öffentlichen Straßenverkehr. Um eine langfristige, systemische Verantwortungsübernahme zur Verkehrsbeobachtung und Reaktion zu realisieren, muss die durchgängige Beherrschbarkeit erwartbarer Situationen ohne Fahrereingriff in der ausgewiesenen Betriebsdomäne sichergestellt werden. Für die Motor- und Bremsenansteuerung des Egofahrzeugs ist dabei die Erfassung und Auswahl relevanter Verkehrsteilnehmer eine entscheidende Herausforderung - insbesondere bei Einschermanövern in die eigene Spur. Sie kann je nach Kritikalität der eintretenden Situation und in Abhängigkeit von der Reaktionsfähigkeit zur Kollision führen. Den technisch-sicherheitsrelevanten Anforderungen zur Realisierung einer fahrerlosen Steuerung stehen den Automobilherstellern dabei u.a. die wirtschaftlichen und normativen Vorgaben gegenüber: Unter Verwendung zahlreicher Steuergeräte und Sensoren, die vorverarbeitete Informationen der erfassten Objekte liefern, muss eine hinreichende Erfüllung der gesetzlichen und marktspezifischen Anforderungen zum Serieneinsatz unter gleichzeitiger Berücksichtigung des Aufwands erfolgen. Ziel der vorliegenden Arbeit ist die Analyse der notwendigen sensorischen Leistungsfähigkeit zur rechtzeitigen Detektion von Spurwechseln anderer Verkehrsteilnehmer in der Betriebsdomäne einer hochautomatisierten Fahrfunktion zur Ermöglichung einer kollisionsvermeidenden Bremsreaktion. Neben der Darstellung der spezifischen Anforderungen dieser Assistenzstufe im Vergleich zu in Serie befindlichen Systemen wird im ersten Schritt die menschliche Leistungsfähigkeit aus zwei Simulatorstudien bestimmt, um eine Vergleichbarkeit der Risikobilanz für die nachfolgenden Modelle zu ermöglichen. Im nächsten Schritt werden aus den analysierten Eigenschaften der Spurwechselcharakteristik, den Normen zur Straßenanlage und den Bewegungen des sensortragenden Egofahrzeugs die Anforderungen an den sensorisch abzudeckenden Merkmalsraum formuliert. Unter Zuhilfenahme einer existierenden, algorithmischen Modellierung mittels Bayesschen Netzen können die sensorischen Daten zur Erkennung des Spurwechselvorgangs probabilistisch überführt werden. Die Parametrierung des Modells wird im Umfang dieser Arbeit unter Einbezug von Realdaten maschinell trainiert und eine Steigerung der Sensitivität ermöglicht. Für die individuellen, fehlerbehafteten sensorischen Eingangsgrößen wird folglich die Eignung im Gesamtkontext der Spurwechselerkennung simulativ untersucht und in Feldversuchen mit übergeordneter Genauigkeit bewertet. Dabei wird abschließend der für den Automobileinsatz bestimmte und einführend vorgestellte Laserscanner Ibeo ScaLa evaluiert. Die Bewertung der ermittelten Genauigkeiten der Objektdetektion sowie der bereitgestellten Fehlerschätzung erfolgen in Bezug zur erwarteten Risikobilanz des hochautomatisierten Fahrens. Als Ergebnis dieser Arbeit kann für die Spurwechseldetektion anderer Verkehrsteilnehmer neben der ermittelten Reaktionsleistung menschlicher Fahrer auch die damit verbundene, weitreichende Anforderungserfüllung für den betrachteten Laserscanner attestiert werden. Die in Extremfällen fehlende Abdeckung im Randbereich des Sichtfeldes lässt sich durch einfache Erweiterungen in der Fahrstrategie der hochautomatisierten Betriebsdomäne beherrschen. Die experimentell ermittelten Gütemaße erlauben eine Detektion der erwartbaren Spurwechsel bis zu einer durch das verbesserte Modell limitierten Dynamikgrenze. Kollisionen können bei kritischen Spurwechseln bis zu dieser Einschränkung vermieden werden.

Autonomous vehicles have gained much importance over the last decade owing to their promising capabilities like improvement in overall traffic flow, reduction in pollution and elimination of human errors. However, when it comes to long-distance transportation or working in complex isolated environments like mines, various factors such as safety, fuel efficiency, transportation cost, robustness, and accuracy become very critical. This thesis, developed at the Connected and Autonomous Systems department of Scania AB in association with KTH, focuses on addressing the issues related to fuel efficiency, robustness and accuracy of an autonomous heavy-duty truck used for mining applications. First, in order to improve the state prediction capabilities of the simulation model, a comparative analysis of two dynamic bicycle models was performed. The first model used the empirical PAC2002 Magic Formula (MF) tyre model to generate the tyre forces, and the latter used a piece-wise Linear approximation of the former. On top of that, in order to account for the nonlinearities and time delays in the lateral direction, the steering dynamic equations were empirically derived and cascaded to the vehicle model. The fidelity of these models was tested against real experimental logs, and the best vehicle model was selected by striking a balance between accuracy and computational efficiency. The Dynamic bicycle model with piece-wise Linear approximation of tyre forces proved to tick-all-the-boxes by providing accurate state predictions within the acceptable error range and handling lateral accelerations up to 4 m/s2. Also, this model proved to be six times more computationally efficient than the industry-standard PAC2002 tyre model. Furthermore, in order to ensure smooth and accurate driving, several Model Predictive Control (MPC) formulations were tested on clothoid-based Single Lane Change (SLC), Double Lane Change (DLC) and Truncated Slalom trajectories with added disturbances in the initial position, heading and velocities. A linear time-varying Spatial error MPC is proposed, which provides a link between spatial-domain and time-domain analysis. This proposed controller proved to be a perfect balance between fuel efficiency which was achieved by minimising braking and acceleration sequences and offset-free tracking along with ensuring that the truck reached its destination within the stipulated time irrespective of the added disturbances. Lastly, a comparative analysis between various Prediction-Simulation model pairs was made, and the best pair was selected in terms of its robustness to parameter changes, simplicity, computational efficiency and accuracy. / Under det senaste årtiondet har utveckling av autonoma fordon blivit allt viktigare på grund av de stora möjligheterna till förbättringar av trafikflöden, minskade utsläpp av föroreningar och eliminering av mänskliga fel. När det gäller långdistanstransporter eller komplexa isolerade miljöer så som gruvor blir faktorer som bränsleeffektivitet, transportkostnad, robusthet och noggrannhet mycket viktiga. Detta examensarbete utvecklat vid avdelningen Connected and Autonomous Systems på Scania i samarbete med KTH fokuserar på frågor gällande bränsleeffektivitet, robusthet och exakthet hos en autonom tung lastbil i gruvmiljö. För att förbättra simuleringsmodellens tillståndsprediktioner, genomfördes en jämförande analys av två dynamiska fordonsmodeller. Den första modellen använde den empiriska däckmodellen PAC2002 Magic Formula (MF) för att approximera däckkrafterna, och den andra använde en stegvis linjär approximation av samma däckmodell. För att ta hänsyn till ickelinjäriteter och laterala tidsfördröjningar inkluderades empiriskt identifierade styrdynamiksekvationer i fordonsmodellen. Modellerna verifierades mot verkliga mätdata från fordon. Den bästa fordonsmodellen valdes genom att hitta en balans mellan noggrannhet och beräkningseffektivitet. Den Dynamiska fordonsmodellen med stegvis linjär approximation av däckkrafter visade goda resultat genom att ge noggranna tillståndsprediktioner inom det acceptabla felområdet och hantera sidoacceleration upp till 4 m/s2 . Den här modellen visade sig också vara sex gånger effektivare än PAC2002-däckmodellen. v För att säkerställa mjuk och korrekt körning testades flera MPC varianter på klotoidbaserade trajektorier av filbyte SLC, dubbelt filbyte DLC och slalom. Störningar i position, riktining och hastighet lades till startpositionen. En MPC med straff på rumslig avvikelse föreslås, vilket ger en länk mellan rumsdomän och tidsdomän. Den föreslagna regleringen visade sig vara en perfekt balans mellan bränsleeffektivitet, genom att minimering av broms- och accelerationssekvenser, och felminimering samtidigt som lastbilen nådde sin destination inom den föreskrivna tiden oberoende av de extra störningarna. Slutligen gjordes en jämförande analys mellan olika kombinationer av simulerings- och prediktionsmodell och den bästa kombinationen valdes med avseende på dess robusthet mot parameterändringar, enkelhet, beräkningseffektivitet och noggrannhet.