With the advancement of autonomous driving research, 3D detection based on LiDAR point cloud has gradually become one of the top research topics in the field of artificial intelligence. Compared with RGB cameras, LiDAR point cloud can provide depth information, while RGB images can provide denser resolution. Features from LiDAR and cameras are considered to be complementary. However, due to the sparsity of the LiDAR point clouds, a dense and accurate RGB/3D projective relationship is difficult to establish especially for distant scene points. Recent works try to solve this problem by designing a network that learns missing points or dense point density distribution to compensate for the sparsity of the LiDAR point cloud. During the master’s exploration, we consider addressing this problem from two aspects. The first is to design a GAN(Generative Adversarial Network)-based module to reconstruct point clouds, and the second is to apply regional point cloud enhancement based on motion maps. In the first aspect, we propose to use an imagine-and-locate process, called UYI. The objective of this module is to improve the point cloud quality and is independent of the detection stage used for inference. We accomplish this task through a GAN-based cross-modality module that uses image as input to infer a dense LiDAR shape. In another aspect, inspired by the attention mechanism of human eyes, we use motion maps to perform random augmentation on point clouds in a targeted manner named motion map-assisted enhancement, MAE. Boosted by our UYI and MAE module, our experiments show a significant performance improvement in all tested baseline models. In fact, benefiting from the plug-and-play characteristics of our module, we were able to push the performance of the existing state-of-the-art model to a new height. Our method not only has made great progress in the detection performance of vehicle objects but also achieved an even bigger leap forward in the pedestrian category. In future research, we will continue to explore the feasibility of spatio-temporal correlation methods in 3D detection, and 3D detection related to motion information extraction could be a promising direction.

Traffic congestion and safety has become a major issue in the modern world's commute. Congestion has been causing people to travel billions of hours more and to purchase billions of gallons of fuel extra which account to congestion cost of billions of dollars. Autonomous driving vehicles have been one solution to this problem because of their huge impact on efficiency, pollution, and human safety. Also, extensive research has been carried out on control design of vehicular platoons because a further improvement in traffic throughput while not compromising the safety is possible when the vehicles in the platoon are provided with better predictive abilities. Motion control is a key area of autonomous driving research that handles moving parts of vehicles in a deliberate and controlled manner. A widely worked on problem in motion control concerned with time parameterized reference tracking is trajectory tracking. Having an efficient and effective tracking algorithm embedded in the autonomous driving system is the key for better performance in terms of resources consumed and tracking error. Many tracking control algorithms in literature rely on an accurate model of the vehicle and often, it can be an intimidating task to come up with an accurate model taking into consideration various conditions like friction, heat effects, ageing processes etc. And typically, control algorithms rely on periodic execution of the tasks that update the control actions, but such updates might not be required, which result in unnecessary actions that waste resources. The main focus of this work is to design an intermittent model-free optimal control algorithm in order to enable autonomous vehicles to track trajectories at high-speeds. To obtain a solution which is model-free, a Q-learning setup with an actor-network to approximate the optimal intermittent controller and a critic network to approximate the optimal cost, resulting in the appropriate tuning laws is considered. / Master of Science / A risen research effort in the area of autonomous vehicles has been witnessed in the past few decades because these systems improve safety, comfort, transport time and energy consumption which are some of the main issues humans are facing in the modern world’s highway systems. Systems like emergency braking, automatic parking, blind angle vehicle detection are creating a safer driving environment in populated areas. Advanced driver assistance systems (ADAS) are what such kind of systems are known as. An extension of these partially automated ADAS are vehicles with fully automated driving abilities, which are able to drive by themselves without any human involvement. An extensively proposed approach for making traffic throughput more efficient on existing highways is to assemble autonomous vehicles into platoons. Small intervehicle spacing and many vehicles constituting each platoon formation improve the traffic throughput significantly. Lately, the advancements in computational capabilities, in terms of both algorithms and hardware, communications, and navigation and sensing devices contributed a lot to the development of autonomous systems (both single and multiagent) that operate with high reliability in uncertain/dynamic operating conditions and environments. Motion control is an important area in the autonomous vehicles research. Trajectory-tracking is a widely studied motion control scenario which is about designing control laws that force a system to follow some time-dependent reference path and it is important to have an effective and efficient trajectory-tracking control law in an autonomous vehicle to reduce the resources consumed and tracking error. The goal of this work is to design an intermittent model-free trajectory tracking control algorithm where there is no need of any mathematical model of the vehicle system being controlled and which can reduce the controller updates by allowing the system to evolve in an open loop fashion and close the loop only when an user defined triggering condition is satisfied. The approach is energy efficient in that the control updates are limited to instances when they are needed rather than unnecessary periodic updates. Q-learning which is a model-free reinforcement learning technique is used in the trajectory tracking motion control algorithm to make the vehicles track their respective reference trajectories without any requirement of their motion model, the knowledge of which is generally needed when dealing with a motion control problem. The testing of the designed algorithm in simulations and experiments is presented in this work. The study and development of a vehicle platform in order to perform the experiments is also discussed. Different motion control and sensing techniques are presented and used. The vehicle platform is shown to track a reference trajectory autonomously without any human intervention, both in simulations and experiments, proving the effectiveness of the proposed algorithm.

<p dir="ltr">The growth of electric vehicles (EVs) offers several benefits for air quality improvement and emissions reduction. Nonetheless, EVs also pose several challenges in the area of highway transportation. These barriers are related to the limitations of EV technology, particularly the charge duration and speed of battery recharging, which translate to vehicle range anxiety for EV users. A promising solution to these concerns is V2V DWC technology (Vehicle to Vehicle Dynamic Wireless Charging), particularly mobile energy disseminators (MEDs). The MED is mounted on a large vehicle or truck that charges all participating EVs within a specified locus from the MED. However, current research on MEDs offers solutions that are widely considered impractical for deployment, particularly in urban environments where range anxiety is common. Acknowledging such gap in the literature, this thesis proposes a comprehensive methodological framework for optimal MED deployment decisions. In the first component of the framework, a practical system, termed “ChargingEnv” is developed using reinforcement learning (RL). ChargingEnv simulates the highway environment, which consists of streams of EVs and an MED. The simulation accounts for a possible misalignment of the charging panel and incorporates a realistic EV battery model. The second component of the framework uses multiple deep RL benchmark models that are trained in “ChargingEnv” to maximize EV service quality within limited charging resource constraints. In this study, numerical experiments were conducted to demonstrate the MED deployment decision framework’s efficacy. The findings indicate that the framework’s trained model can substantially improve EV travel range and alleviate battery depletion concerns. This could serve as a vital tool that allows public-sector road agencies or private-sector commercial entities to efficiently orchestrate MED deployments to maximize service cost-effectiveness.</p>

In recent years, autonomous driving has become an increasingly practical technology. With state-of-the-art computer and sensor engineering, autonomous vehicles may be produced and widely used for travel and logistics in the near future. They have great potential to reduce traffic accidents, improve transportation efficiency, and release people from driving tasks while commuting. Researchers have built autonomous vehicles that can drive on public roads and handle normal surrounding traffic and obstacles. However, in situations like lane changing and merging, the autonomous vehicle faces the challenge of performing smooth interaction with human-driven vehicles. To do this, autonomous vehicle intelligence still needs to be improved so that it can better understand and react to other human drivers on the road. In this thesis, we argue for the importance of implementing ”socially cooperative driving”, which is an integral part of everyday human driving, in autonomous vehicles. An intention-integrated Prediction- and Cost function-Based algorithm (iPCB) framework is proposed to enable an autonomous vehicles to perform cooperative social behaviors. We also propose a behavioral planning framework to enable the socially cooperative behaviors with the iPCB algorithm. The new architecture is implemented in an autonomous vehicle and can coordinate the existing Adaptive Cruise Control (ACC) and Lane Centering interface to perform socially cooperative behaviors. The algorithm has been tested in over 500 entrance ramp and lane change scenarios on public roads in multiple cities in the US and over 10; 000 in simulated case and statistical testing. Results show that the proposed algorithm and framework for autonomous vehicle improves the performance of autonomous lane change and entrance ramp handling. Compared with rule-based algorithms that were previously developed on an autonomous vehicle for these scenarios, over 95% of potentially unsafe situations are avoided.

Vehicle-to-Vehicle communication is the key technology for achieving increased perception for automated vehicles where the communication allows virtual sensing with the use of sensors placed in other vehicles. In addition, this technology also allows recognising objects that are out-of-sight. This thesis presents a Trust System that allows a vehicle to make more reliable and robust decisions. The system evaluates the current situation and generates a Trust Index indicating the level of trust in the environment, the ego vehicle, and the other vehicles. Current research focuses on securing the communication between the vehicles themselves, but does not verify the content of the received data on a system level. The proposed Trust System evaluates the received data according to sensor accuracy, behaviour of other vehicles, and the perception of the local environment. The results show that the proposed method is capable of correctly identifying various situations and discusses how the Trust Index can be used to make more robust decisions.

Trajectory planning is an important component of autonomous driving. It takes the result of route-level navigation plan and generates the motion-level commands that steer an autonomous passenger vehicle (APV). Prior work on solving this problem uses either a sampling-based or optimization-based trajectory planner, accompanied by some high-level rule generation components.

This thesis is concerned with the problem of situational awareness in autonomous vehicles. In this context, situational awareness refers to the ability of an autonomous vehicle to perceive the road layout ahead, interpret the implied semantics and gain an awareness of its surrounding - thus reading the road ahead. Autonomous vehicles require a high level of situational awareness in order to operate safely and efficiently in real-world dynamic environments. A system is therefore needed that is able to model the expected road layout in terms of semantics, both under normal and roadwork conditions. This thesis takes a three-pronged approach to this problem: Firstly, we consider reading the road surface. This is formulated in terms of probabilistic road marking classification and interpretation. We then derive the road boundaries using only a 2D laser and algorithms based on geometric priors from Highway Traffic Engineering principles. Secondly, we consider reading the road scene. Here, we formulate a roadwork scene recognition framework based on opponent colour vision in humans. Finally, we provide a data representation for situational awareness that unifies reading the road surface and reading the road scene. This thesis therefore frames situational awareness in autonomous vehicles in terms of both static and dynamic road semantics - and detailed formulations and algorithms are discussed. We test our algorithms on several benchmarking datasets collected using our autonomous vehicle on both rural and urban roads. The results illustrate that our road boundary estimation, road marking classification, and roadwork scene recognition frameworks allow autonomous vehicles to truly and meaningfully read the semantics of the road ahead, thus gaining a valuable sense of situational awareness even at challenging layouts, roadwork sites, and along unknown roadways.

In order to enable more widespread application of robots, we are required to reduce the human effort for the introduction of existing robotic platforms to new environments and tasks. In this thesis, we identify three complementary strategies to address this challenge, via the use of imitation learning, domain adaptation, and transfer learning based on simulations. The overall work strives to reduce the effort of generating training data by employing inexpensively obtainable labels and by transferring information between different domains with deviating underlying properties. Imitation learning enables a straightforward way for untrained personnel to teach robots to perform tasks by providing demonstrations, which represent a comparably inexpensive source of supervision. We develop a scalable approach to identify the preferences underlying demonstration data via the framework of inverse reinforcement learning. The method enables integration of the extracted preferences as cost maps into existing motion planning systems. We further incorporate prior domain knowledge and demonstrate that the approach outperforms the baselines including manually crafted cost functions. In addition to employing low-cost labels from demonstration, we investigate the adaptation of models to domains without available supervisory information. Specifically, the challenge of appearance changes in outdoor robotics such as illumination and weather shifts is addressed using an adversarial domain adaptation approach. A principal advantage of the method over prior work is the straightforwardness of adapting arbitrary, state-of-the-art neural network architectures. Finally, we demonstrate performance benefits of the method for semantic segmentation of drivable terrain. Our last contribution focuses on simulation to real world transfer learning, where the characteristic differences are not only regarding the visual appearance but the underlying system dynamics. Our work aims at parallel training in both systems and mutual guidance via auxiliary alignment rewards to accelerate training for real world systems. The approach is shown to outperform various baselines as well as a unilateral alignment variant.

L'automatisation de la conduite est un processus en cours qui est en train de changer radicalement la façon dont les gens voyagent et passent du temps dans leur voiture pendant leurs déplacements. Les véhicules conditionnellement automatisés libèrent les conducteurs humains de la surveillance et de la supervision du système et de l'environnement de conduite, leur permettant d'effectuer des activités secondaires pendant la conduite, mais requièrent qu’ils puissent reprendre la tâche de conduite si nécessaire. Pour les conducteurs, il est essentiel de comprendre les capacités et les limites du système, d’en reconnaître les notifications et d'interagir de manière adéquate avec le véhicule pour assurer leur propre sécurité et celle des autres usagers de la route. À cause de la diversité des situations de conduite que le conducteur peut rencontrer, les programmes traditionnels de formation peuvent ne pas être suffisants pour assurer une compréhension efficace de l'interaction entre le conducteur humain et le véhicule pendant les transitions de contrôle. Il est donc nécessaire de permettre aux conducteurs de vivre ces situations avant leur première utilisation du véhicule. Dans ce contexte, la Réalité Mixte constitue un outil d'apprentissage et d'évaluation des compétences potentiellement efficace qui permettrait aux conducteurs de se familiariser avec le véhicule automatisé et d'interagir avec le nouvel équipement dans un environnement sans risque. Si jusqu'à il y a quelques années, les plates-formes de Réalité Mixte étaient destinées à un public de niche, la démocratisation et la diffusion à grande échelle des dispositifs immersifs ont rendu leur adoption plus accessible en termes de coût, de facilité de mise en œuvre et de configuration. L'objectif de cette thèse est d'étudier le rôle de la réalité mixte dans l'acquisition de compétences pour l'interaction d'un conducteur avec un véhicule conditionnellement automatisé. En particulier, nous avons exploré le rôle de l'immersion dans le continuum de la réalité mixte en étudiant différentes combinaisons d'espaces de visualisation et de manipulation et la correspondance entre le monde virtuel et le monde réel. Du fait des contraintes industrielles, nous avons limité les candidats possibles à des systèmes légers portables, peu chers et facilement accessibles; et avons analysé l’impact des incohérences sensorimotrices que ces systèmes peuvent provoquer sur la réalisation des activités dans l’environnement virtuel. À partir de ces analyses, nous avons conçu un programme de formation visant l'acquisition des compétences, des règles et des connaissances nécessaires à l'utilisation d'un véhicule conditionnellement automatisé. Nous avons proposé des scénarios routiers simulés de plus en plus complexes pour permettre aux apprenants d’interagir avec ce type de véhicules dans différentes situations de conduite. Des études expérimentales ont été menées afin de déterminer l'impact de l'immersion sur l'apprentissage, la pertinence du programme de formation conçu et, à plus grande échelle, de valider l'efficacité de l'ensemble des plateformes de formation par des mesures subjectives et objectives. Le transfert de compétences de l'environnement de formation à la situation réelle a été évalué par des essais sur simulateurs de conduite haut de gamme et sur des véhicules réels sur la voie publique. / Driving automation is an ongoing process that is radically changing how people travel and spend time in their cars during journeys. Conditionally automated vehicles free human drivers from the monitoring and supervision of the system and driving environment, allowing them to perform secondary activities during automated driving, but requiring them to resume the driving task if necessary. For the drivers, understanding the system’s capabilities and limits, recognizing the system’s notifications, and interacting with the vehicle in the appropriate way is crucial to ensuring their own safety and that of other road users. Because of the variety of unfamiliar driving situations that the driver may encounter, traditional handover and training programs may not be sufficient to ensure an effective understanding of the interaction between the human driver and the vehicle during transitions of control. Thus, there is the need to let drivers experience these situations before their first ride. In this context, Mixed Reality provides potentially valuable learning and skill assessment tools which would allow drivers to familiarize themselves with the automated vehicle and interact with the novel equipment involved in a risk-free environment. If until a few years ago these platforms were destined to a niche audience, the democratization and the large-scale spread of immersive devices since then has made their adoption more accessible in terms of cost, ease of implementation, and setup. The objective of this thesis is to investigate the role of Mixed Reality in the acquisition of competences needed for a driver’s interaction with a conditionally automated vehicle. In particular, we explored the role of immersion along the Mixed Reality continuum by investigating different combinations of visualization and manipulation spaces and the correspondence between the virtual and the real world. For industrial constraints, we restricted the possible candidates to light systems that are portable, cost-effective and accessible; we thus analyzed the impact of the sensorimotor incoherences that these systems may cause on the execution of tasks in the virtual environment. Starting from these analyses, we designed a training program aimed at the acquisition of skills, rules and knowledge necessary to operate a conditionally automated vehicle. In addition, we proposed simulated road scenarios with increasing complexity to suggest what it feels like to be a driver at this level of automation in different driving situations. Experimental user studies were conducted in order to determine the impact of immersion on learning and the pertinence of the designed training program and, on a larger scale, to validate the effectiveness of the entire training platform with self-reported and objective measures. Furthermore, the transfer of skills from the training environment to the real situation was assessed with test drives using both high-end driving simulators and actual vehicles on public roads.

When introducing autonomous functionality in personal vehicles the ability to control the quality of the ride is transferred from the driver to the vehicle control system. In this context, a reference method for quantifying ride quality may be a useful tool in the development and tuning process. This master’s thesis investigates whether general quantitative measures of ride quality can be of value in the tuning of motion controllers for autonomous vehicles. A set of tools is built for a specific case study, analysing a lateral path planning algorithm, based on a finite horizon linear quadratic tracking controller, and how its tuning affects ride quality performance. A graphical user interface is built, with functionality for frequency domainanalysis of the path planning algorithm, individually and in combination with the remaining lateral control system, as well as ride quality evaluation based on lateral acceleration data, from logged test runs and simulation results. In addition, a simulation environment for the lateral control system is modified to be used in combination with the evaluation tool. Results of the case study indicate a measurable difference in ride quality performance when comparing manual and autonomous driving with the current implementation. Attempts were made to improve ride quality by re-tuning the path planning algorithm but little or no improvement from the previous tuning was made. The work has recognized the potential of using ride quality measures in the development and tuning process for autonomous vehicles as well as devising a tuning strategy incorporating frequency analysis and ride quality evaluation through simulation for the lateral control system. To further increase ride quality performance via the path planning algorithm an altered controller structure, such as a frequency weighted linear quadratic controller is suggested.