Search

In this paper, a new framework for path tracking is proposed through learning to drive like human beings. Firstly, the imitation algorithm (behavior cloning) is adopted to initialize the deep reinforcement learning (DRL) algorithm through learning the professional drivers’ experience. Secondly, a continuous, deterministic, model free deep reinforcement learning algorithm is adopted to optimize our DRL model on line through trial and error. By combining behavior cloning and deep reinforcement learning algorithms, the DRL model can learn an effective policy quickly for path tracking using some easy-to-measure vehicle state parameters and environment information as inputs. Actor-Critic structure is adopted in the DRL algorithm. In order to speed up the convergence rate of the DRL model and improve the learning effect, we propose a dual actor networks structure for the two different action outputs (steering wheel angle and vehicle speed), and a chief critic network is built to guide the updating process of dual actor networks at the same time. Based on this dual actor networks structure, we can pick out some more important state information as state inputs for different action outputs. Besides, a kind of reward mechanism is also presented for autonomous driving. Finally, simulation training and experiment test are carried out, and the results confirm that the framework proposed in this paper is more than data efficient than the original algorithm, and the trained DRL model can track the reference path with accuracy and has the generalization ability for different roads. [ABSTRACT FROM AUTHOR]

This paper studies the relative pose problem for autonomous vehicles driving in highly dynamic and possibly cluttered environments. This is a challenging scenario due to the existence of multiple, large, and independently moving objects in the environment, which often leads to an excessive portion of outliers and results in erroneous motion estimation. Existing algorithms cannot cope with such situations well. This paper proposes a new algorithm for relative pose estimation using a multi-camera system with multiple non-overlapping cameras. The method works robustly even when the number of outliers is overwhelming. By exploiting specific prior knowledge of the autonomous driving scene, we have developed an efficient 4-point algorithm for multi-camera relative pose estimation, which admits analytic solutions by solving a polynomial root-finding equation, and runs extremely fast (at about 0.5 $\mu \text{s}$ per root). When the solver is used in combination with a new random sample consensus sampling scheme by exploiting the conjugate motion constraint, we are able to quickly prune unpromising hypotheses and significantly improve the chance of finding inliers. Experiments on synthetic data have validated the performance of the proposed algorithm. Tests on real data further confirm the method's practical relevance. [ABSTRACT FROM AUTHOR]

The future of autonomous driving is slowly approaching, but there are still many steps to take before it can become a reality. It is crucial to pay attention to road infrastructure, because without it, intelligent vehicles will not be able to operate reliably, and it will never be possible to dispense of driver’s control. This paper presents the work carried out for the detection of road markings damage using computer vision techniques. This is a complex task for which there are currently not many papers and large image sets in the literature. This study uses images from the public Road Damage Detection dataset for the D44 defect and also provides 971 new labelled images for Spanish roads. For this purpose, three detectors based on deep learning architectures (Faster RCNN, SDD and EfficientDet) have been used and single-source and mixed-source models have been studied to find the model that best fits the target images. Finally, F1-score values reaching 0.929 and 0.934 have been obtained for Japanese and Spanish images respectively which improve the state-of-the-art results by 25%. It can be concluded that the results of this study are promising, although the collection of many more images will be necessary for the scientific community to continue advancing in the future in this field of research. [ABSTRACT FROM AUTHOR]

The integration of global navigation satellite system (GNSS) and inertial navigation system (INS) has been widely studied in the past decades. The traditional Kalman filtering method, a procedure that leads to historical states and observations loss, can estimate fast by limiting the update to the latest state. Furthermore, the performance of loosely coupled GNSS/INS navigation integration depends largely on environmental conditions and sensor costs. In cities, GNSS signals may be interrupted due to the influence of obstructions and moving objects, which leads to the discontinuity of GNSS/INS navigation. Therefore, in order to overcome these difficulties an auto regressive integrated moving average (ARIMA) auxiliary model based on time sequence is proposed in this paper, which makes use of the data before interruption to predict the GNSS measurements during the interruption period and makes up for the data gap. In addition, the INS and GNSS measurements are loosely coupled using the most advanced factor graph model. Compared to the filtering algorithm, all previous measurements are used for state estimation through the factor graph framework. In this paper, the experiment is based on the dataset of Tokyo which is a typical challenging city canyon. Experiments showed that, compared with the traditional GNSS/INS integration method, the proposed method can provide a higher precision and a series of continuous navigation results even when the GNSS measurement data was interrupted. [ABSTRACT FROM AUTHOR]

End-to-end steering estimation is one of the important deep regression tasks. However, driving datasets are always imbalanced on the distribution of the steering value, which makes end-to-end learning models perform bad steering estimation accuracy, especially for the sharp steering value estimation, which finally, leads to an imbalanced training problem. In this paper, the essential reason for the imbalanced training problem for the deep regression task is first revealed as the gradient disharmony. To hedge the gradient disharmony, a general gradient balanced mechanism for the deep regression task is proposed in this paper. Meanwhile, the general gradient balanced mechanism is applied for the steering estimation task. According to the steering distribution, a novel Adaptive Gradient Balanced Mechanism (AGBM) is proposed to hedge the gradient disharmony based on the Laplace distribution with the general gradient balanced mechanism theory. Further, a new loss function embedding with AGBM is proposed to balance the gradient disharmony for the training process. AGBM makes the process of gradient hedging with adaptive style without parameters finetuning. Based on five driving datasets, four experiments are conducted for demonstrations. The experiment results show the AGBM enables the end-to-end models to perform the best accuracy of the steering estimation, including the sharp steering samples. Meanwhile, AGBM is generalized with different end-to-end learning models. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new state representation method, called encoding sum and concatenation (ESC), to describe the environment observation for decision-making in autonomous driving. Unlike existing state representation methods, ESC is applicable to the situation where the number of surrounding vehicles is variable and eliminates the need for manually pre-designed sorting rules, leading to higher representation ability and generality. The proposed ESC method introduces a feature neural network (NN) to encode the real-valued feature of each surrounding vehicle into an encoding vector, and then adds these vectors up to obtain the representation vector of the set of surrounding vehicles. Then, a fixed-dimensional and permutation-invariance state representation can be obtained by concatenating the set representation with other variables, such as indicators of the ego vehicle and road. By introducing the sum-of-power mapping, this paper has further proved that the injectivity of the ESC state representation can be guaranteed if the output dimension of the feature NN is greater than the number of variables of all surrounding vehicles. This means that the ESC representation can be used to describe the environment and taken as the inputs of learning-based policy functions. Experiments demonstrate that compared with the fixed-permutation representation method, the policy learning accuracy based on ESC representation is improved by 62.2%. [ABSTRACT FROM AUTHOR]

Identifying and evaluating the potential risks in the surrounding environment is critical for intelligent vehicles’ safety and user experience. This paper provides a comprehensive overview of the state-of-the-art research on the surrounding vehicles’ lane change maneuver prediction and detection. First, various driver behavior modeling and classification methods are reviewed and analyzed, which gives a general understanding of what the lane change maneuver is and how to predict or detect the lane change maneuver. Next, the primary sensing devices equipped on intelligent vehicles and their impacts on lane change inference systems are discussed. Then, a series of representative research works in recent years are selected, introduced, and compared regarding their input feature selection, inference algorithms, and performance evaluation methods. Finally, some potential future research directions are proposed. This paper aims to help the relevant researchers and institutions summarize the current studies on the surrounding vehicles’ lane change maneuver inference and recognize its future development directions. [ABSTRACT FROM AUTHOR]

Autonomous Driving (AD) introduces dramatic changes to the way we travel. This emerging technology has the potential to impact the transportation sector across a wide array of categories including safety, efficiency, congestion, legislation, and travel behavior. In this survey, we review the main issues involved in AD as discussed in the literature, and shed light on topics that we consider requiring further development based on Google Scholar and the Web of Science (WoS) data base. The paper also provides the results of research trends related to Autonomous Driving based on analysis of the number of search results listed in google trends. According to our research, the fields of Vehicle-to-Vehicle (V2V) and Vehicle-to-Cloud (V2C) networking are of higher interest due to the technological gaps and standardization processes. In addition, cyber and security research is in acceleration due to its importance. [ABSTRACT FROM AUTHOR]

Most of the recent automated driving systems assume the accurate functioning of localization. Unanticipated errors cause localization failures and result in failures in automated driving. An exact localization failure detection is necessary to ensure safety in automated driving; however, detection of the localization failures is challenging because sensor measurement is assumed to be independent of each other in the localization process. Owing to the assumption, the entire relation of the sensor measurement is ignored. Consequently, it is difficult to recognize the misalignment between the sensor measurement and the map when partial sensor measurement overlaps with the map. This paper proposes a method for the detection of localization failures using Markov random fields with fully connected latent variables. The full connection enables to take the entire relation into account and contributes to the exact misalignment recognition. Additionally, this paper presents localization failure probability calculation and efficient distance field representation methods. We evaluate the proposed method using two types of datasets. The first dataset is the SemanticKITTI dataset, whereby four methods are compared with the proposed method. The comparison results reveal that the proposed method achieves the most accurate failure detection. The second dataset is created based on log data acquired from the demonstrations that we conducted in Japanese public roads. The dataset includes several localization failure scenes. We apply the failure detection methods to the dataset and confirm that the proposed method achieves exact and immediate failure detection. [ABSTRACT FROM AUTHOR]

Semantic segmentation plays a critical role in scene understanding for self-driving vehicles. A line of efforts has proven that global context matters in urban scene segmentation due to massive scale changes. However, we find that existing methods suffer from local ambiguities when dissipating continuous local context, i.e. scrambling to a huge receptive field of global cues by coarse pooling. To this end, this paper proposes a new Context Aggregation Module (CAM) that consists of two primary components: context encoding using no coarse pooling but encoder-decoders with appropriate sampling scales and gated fusion that extends gate attention mechanism to balance different-scale context during feature fusion. Weeding out coarse pooling and applying the encoder-decoder inherits the merits of exploring global context while avoiding the drawback of losing local contextual continuity. We then construct a Context Aggregation Network (CANet) and conduct extensive evaluations on challenging autonomous driving benchmarks of Cityscapes, CamVid and BDD100K. Consistently improved results evidence the effectiveness. Notably, we attain competitive mIoU 82.7% on Cityscapes and optimal mIoU 80.5% on CamVid. [ABSTRACT FROM AUTHOR]

This paper proposes a real-time multi-scale semantic segmentation network (MsNet). MsNet is a combination of our novel multi-scale fusion with matching attention model (MFMA) as the decoding network and the network searched by asymptotic neural architecture search (ANAS) or MobileNetV3 as the encoding network. The MFMA not only extracts low level spatial features from multi-scale inputs but also decodes the contextual features extracted by ANAS. Specifically, considering the advantages and disadvantages of the addition fusion and concatenation fusion, we design multi-scale fusion (MF) that balances speed and accuracy. Then we creatively design two matching attention mechanisms (MA), including matching attention with low calculation (MALC) mechanism and matching attention with strong global context modeling (MASG) mechanism, to match varying resolutions and information of features at different levels of a network. Besides, the ANAS performs the deep neural network search by employing an asymptotic method and provide an efficient encoding network for MsNet, releasing researchers from those tedious mechanical trials. Through extensive experiments, we prove that MFMA, which can be applied to numerous recognition tasks, possesses excellent decoding ability. And we demonstrate the effectiveness and necessity of implementing the “matching” attention mechanism. Finally, the proposed two versions, MsNet_ANAS and MsNet_M achieve a new state-of-the-art trade-off between accuracy and speed on the CamVid and Cityscapes datasets. More remarkably, on the Nvidia Tesla V100 GPU, our MsNet_ANAS achieves 74.1% mIoU with the speed of 184.2 FPS on the CamVid while 72.9% mIoU with the speed of 119.9 FPS on the Cityscapes. [ABSTRACT FROM AUTHOR]

The complex traffic and road environment pose considerable challenges to the accuracy, timeliness, and adaptive ability of connected and autonomous vehicles (CAVs) in making driving decisions. This paper uses vehicle collaboration and integrates the adaptive learning capabilities of machine learning and the interpretation capabilities of expert systems (ESs) in a unified architecture to form a hybrid autonomous driving guidance system, which not only solves the “bottleneck” of knowledge acquisition during the construction of expert systems but also solves the “black box” phenomenon of machine learning in the decision-making process. First, an autonomous driving strategy based on deep reinforcement learning (DRL) is proposed for CAVs to make decisions and extract corresponding rules. Next, we design an ES knowledge base expansion method including rule extraction, rule sharing, and rule test. Particularly, vehicular blockchain is adopted to ensure user privacy and data security during the rule-sharing process. Third, hybrid autonomous driving guidance combining ES and machine learning is proposed for CAVs to make accurate and efficient decisions in different driving environments. Once the strategy is well trained, it can effectively guide CAVs to cope with the complex traffic environment. Extensive simulations validate the performance of our proposal in terms of decision-making accuracy, effectiveness, and safety. [ABSTRACT FROM AUTHOR]

Imbalanced training is a challenge in the field of autonomous driving. For the steering estimation task, imbalanced training is the core reason that an end-to-end model cannot estimate sharp steering value well. Inspired by researches on the steering estimation, this paper proposes a novel loss function to train a high-performance end-to-end model for handling the imbalanced training problem, which is named SteeringLoss. Firstly, the imbalanced distribution of the steering value for driving datasets is analyzed, which is similar to the double long-tailed distribution. Secondly, with the feature of distribution, this paper designs a cost-sensitive loss function step by step, the new loss function can improve the impact of the sharp steering value while maintaining the impact of the small steering value. Thirdly, three typical steering estimation models are established with SteeringLoss for demonstration, including the CNN model, the CNN-LSTM model and the 3DCNN-LSTM model. Finally, three experiments are designed for SteeringLoss: Experiment I demonstrates SteeringLoss can avoid imbalanced training problem; Experiment II discusses the finetuning principle of SteeringLoss and gives the basic guideline for using SteeringLoss; Experiment III shows the results on different typical end-to-end steering estimation models, which shows effectiveness of SteeringLoss for all the models and gives different solutions for the steering estimation. Moreover, the SteeringLoss is suitable for the imbalanced training with similar distributions of datasets besides end-to-end steering estimation, which indicates the potential value of the research on SteeringLoss in the future. [ABSTRACT FROM AUTHOR]

Autonomous vehicles need to solve the road keeping problem and the existing solutions based on reinforcement learning are mainly implemented in the simulators. The key of transferring the well-trained models to the real world is bridging the gaps between the simulator scenarios and the real scenarios. In this paper, we propose a method called ReinforcementDriving which explores navigation skills and trajectories from simulator for full-sized road keeping. Based on the real scenario, a driving simulator is firstly established to train an intelligent driving agent. The well-trained ReinforcementDriving agent is evaluated in a real-world scenario. We compare our work with human driving, optimal control-based tracking methods and other reinforcement learning-based lane following methods. The results demonstrate that the ReinforcementDriving system can effectively achieve lane keeping in a realistic scenario with satisfactory running time and lateral accuracy. [ABSTRACT FROM AUTHOR]

The rapid development of 6G can help to bring autonomous driving closed to the reality. Drivers and passengers will have more time for work and leisure spending in the vehicles, further generating a lot of data requirements. However, edge resources from small base stations are insufficient to match the wide variety of services of the future vehicular networks. Besides, due to the high-speed nature of the vehicles, users have to switch the connections among different base stations, whereas such way will cause external latency during the data request. Therefore, it is vital to enable the local cache of vehicle users to realize the reliable autonomous driving. In this paper, we consider caching the contents in the local cache, small base station, and edge server. In practice, the request preference of some single users may be different from a whole region. To maximize the efficiency of content cache, we design a strategy that uses reinforcement learning algorithm to optimize cache schemes on different devices. The experimental results demonstrate that our strategy can enhance the cache hit ratio by 10%-20% compared with the well-known counterparts. [ABSTRACT FROM AUTHOR]

Although autonomous vehicle technology has evolved significantly in recent years, the navigation of self-driving vehicles in complex scenarios is still an open issue. One of the major challenges in these conditions is safe navigation on roads open to public traffic. The main issue is the interaction of the autonomous vehicle with regular traffic, as the behaviors and intentions of human-driven vehicles are hard to predict and understand. In this paper we propose a strategy to allow an autonomous vehicle to safely cross a multi-lane roundabout. Our approach uses a High-Definition (HD) map to predict at lane level the future situation, harnessing the concept of virtual instances of road users, which is a key concept in anticipating the situation in a roundabout that can be represented by a navigation graph with loops. This paper presents a methodology that uses intervals representing road occupancy by vehicles, with the road being widened to reflect uncertainties in localization. Our method safely avoids collisions and guarantees that no priority constraints are violated during the insertion maneuver. Moreover, the method does not provide an overly cautious insertion policy, i.e., an autonomous vehicle does not wait for a long time before the insertion. The performance of our strategy was evaluated using the SUMO simulation framework. To better evaluate the complexity of the simulation scenario, a highly interactive vehicle flow was generated using real dynamic traffic data from the INTERACTION dataset. We report real tests carried out with an experimental self-driving vehicle on a test circuit. Our results show that this approach is easy to integrate into an embedded system and that it allows roundabouts to be crossed with a high level of safety. [ABSTRACT FROM AUTHOR]

Autonomous vehicles will generate data from the variety of sensors they employ to track the surrounding environment. This data is inherently valuable, as it gives algorithm designers the potential to leverage prior experience in order to improve driving performance over time. This paper uses the lens of autonomous racing to provide an example of how data from previous iterations of driving can be used to improve quantitative metrics of performance. Two complementary algorithms are demonstrated in this paper. The first algorithm uses iterative learning control (ILC) to simultaneously improve lateral and longitudinal tracking of the desired racing trajectory over multiple laps, while the second algorithm is focused on altering the trajectory itself using a search method. When driven experimentally at the limits of handling, the result is a reduction in lap time of nearly 1.4 seconds, a major improvement. [ABSTRACT FROM AUTHOR]

With recent advances in learning algorithms and hardware development, autonomous cars have shown promise when operating in structured environments under good driving conditions. However, for complex, cluttered, and unseen environments with high uncertainty, autonomous driving systems still frequently demonstrate erroneous or unexpected behaviors that could lead to catastrophic outcomes. Autonomous vehicles should ideally adapt to driving conditions; while this can be achieved through multiple routes, it would be beneficial as a first step to be able to characterize driveability in some quantified form. To this end, this paper aims to create a framework for investigating different factors that can impact driveability. Also, one of the main mechanisms to adapt autonomous driving systems to any driving condition is to be able to learn and generalize from representative scenarios. The machine learning algorithms that currently do so learn predominantly in a supervised manner and consequently need sufficient data for robust and efficient learning. Therefore, we also perform a comparative overview of 54 public driving datasets that enable learning and publish this dataset index at https://sites.google.com/view/driveability-survey-datasets. Specifically, we categorize the datasets according to the use cases and highlight the datasets that capture the complicated and hazardous driving conditions, which can be better used for training robust driving models. Furthermore, by discussions of what driving scenarios are not covered by the existing public datasets and what driveability factors need more investigation and data acquisition, this paper aims to encourage both targeted dataset collection and the proposal of novel driveability metrics that enhance the robustness of autonomous cars in adverse environments. [ABSTRACT FROM AUTHOR]

One of the fundamental challenges in the design of perception systems for autonomous vehicles is validating the performance of each algorithm under a comprehensive variety of operating conditions. In the case of vision-based semantic segmentation, there are known issues when encountering new scenarios that are sufficiently different to the training data. In addition, even small variations in environmental conditions, such as illumination and precipitation, can affect the classification performance of the segmentation model. Given the reliance on visual information, these effects often translate into poor semantic pixel classification which can potentially lead to catastrophic consequences when driving autonomously. This paper presents a novel method for analyzing the robustness of semantic segmentation models and provides a number of metrics to evaluate the classification performance over a variety of environmental conditions. The process incorporates an additional sensor (lidar) to automate the process and improve the system integrity, eliminating the need for labor-intensive hand labeling of validation data. The experimental results are presented based on multiple datasets collected at different times of the year with different environmental conditions. We extract the ”Road” class using the lidar to demonstrate the concepts, but this could be extended to other classes with different feature detection algorithms. These results show that the semantic segmentation performance varies depending on the weather, camera parameters, and existence of shadows. The results also demonstrate how the metrics can be used to compare and validate the performance after making improvements to a model, and compare the performance of different networks. [ABSTRACT FROM AUTHOR]

The field of automotive user interfaces has developed rapidly over the last several years. To date, the field has primarily focused on creating user interfaces that promote safe driving, including when the driver is engaged in a secondary task in addition to operating the vehicle. However, researchers now need to prepare for a major change in the automotive domain: the automated driving revolution. The authors argue for a new research agenda that focuses on four challenges for automotive user interfaces: assuring safety in the age of automation, transforming vehicles into places for productivity and play, taking advantage of new mobility options made possible by automated vehicles, while throughout all this preserving user privacy and data security. This article is part of a special issue on smart vehicle spaces. [ABSTRACT FROM PUBLISHER]

GNSS/INS integrated solution has been extensively studied over the past decades. However, its performance relies heavily on environmental conditions and sensor costs. The GNSS positioning can obtain satisfactory performance in the open area. Unfortunately, its accuracy can be severely degraded in a highly urbanized area, due to the notorious multipath effects and none-line-of-sight (NLOS) receptions. As a result, excessive GNSS outliers occur, which causes a huge error in GNSS/INS integration. This paper proposes to apply a fish-eye camera to capture the sky view image to further classify the NLOS and line-of-sight (LOS) measurements. In addition, the raw INS and GNSS measurements are tightly integrated using a state-of-the-art probabilistic factor graph model. Instead of excluding the NLOS receptions, this paper makes use of both the NLOS and LOS measurements by treating them with different weightings. Experiments conducted in typical urban canyons of Hong Kong showed that the proposed method could effectively mitigate the effects of GNSS outliers, and an improved accuracy of GNSS/INS integration was obtained, when compared with the conventional GNSS/INS integration. [ABSTRACT FROM AUTHOR]

The autonomous vehicles (AVs), like that in knight rider, were completely a scientific fiction just a few years ago, but are now already practical with real-world commercial deployments. A salient challenge of AVs, however, is the intensive computing tasks to carry out on board for the real-time traffic detection and driving decision making; this imposes heavy load to AVs due to the limited computing power. To explore more computing power and enable scalable autonomous driving, in this paper, we propose a collaborative task computing scheme for AVs, in which the AVs in proximity dynamically share idle computing power among each other. This, however, raises another fundamental problem on how to incentivize AVs to contribute their computing power and how to fully utilize the pool of group computing power in an optimal way. This paper studies the problem by modeling the issue as a market-based optimal computing resource allocation problem. In specific, we develop a software-defined network (SDN) architecture and consider a star topology where a centered AV outsources its computing tasks to the surrounding AVs for its autonomous driving. A market mechanism is developed in which the surrounding AVs sell their computing power at a cost based on their local idle computing resources. Then, we classify the tasks requested by the centered AV into two types which are task with time to live (TTL) and task without TTL, respectively. With different task types, we define corresponding cost models of the centered AV and formulate them as two minimization problems. The optimal solutions of the problems are achieved to guide the centered AV to wisely allocate computing tasks to surrounding AVs towards minimal cost. Finally, the performance of the proposed scheme is evaluated using simulations, which show that the proposed scheme can result in the guaranteed computing performance yet the lowest costs compared with other conventional schemes. [ABSTRACT FROM AUTHOR]

In this paper, the problem of decision making for autonomous vehicles changing lanes is addressed by formulating multiple games in normal form for pairs of agents. This formulation generates the optimal action for the Ego vehicle at a given state and does not consider global optimality for all agents. The payoff matrices of the games are designed based on a user-defined set of rules. The constant parameters of these payoffs are then adjusted using neural learning to generate optimal behavior among the vehicles. An algorithm integrating deep reinforcement learning and game theory, regarded as Nash Q-learning, is included in the decision-making scheme. The applicability of the proposed method in a lane-changing scenario is tested via simulation. [ABSTRACT FROM AUTHOR]

Robust and adaptive vehicle state estimation and tracking algorithms have become a very important part within the autonomous driving field. The family of interacting multiple model (IMM) filters has shown to provide very effective and accurate state estimation in systems whose behavior patterns change significantly over time. The main reason for the improved performance of IMM filters compared to single model approaches is the mode mixing, which constantly aligns the different models. This paper proposes an innovative way for the mode mixing, when the state-vectors of the models are of different size. The proposed mixing approach consists of two parts: firstly mixing the common states and secondly weighting between original and mixed states based on the model probabilities. Results from artificial simulations and real world measurement setups are shown to demonstrate the validity of the approach. Compared to previously suggested solutions, the proposed approach is more general and the overall complexity of the mode mixing step is reduced, which is the main contribution of the presented paper. [ABSTRACT FROM AUTHOR]

Traffic awareness is the prerequisite of autonomous driving. Given the limitation of on-board sensors (e.g., precision and price), remote measurement from either infrastructure or other vehicles can improve traffic safety. However, the wireless communication carrying the measurement result undergoes fading, noise and interference and has a certain probability of outage. When the communication fails, the vehicle state can only be predicted by Bayesian filtering with a low precision. Higher communication resource utilization (e.g., transmission power) reduces the outage probability and hence results in an improved estimation precision. The power control subject to an estimate variance constraint is a difficult problem due to the complicated mapping from transmit power to vehicle-state estimate variance. In this paper, we develop an estimator consisting of several Kalman filters (KFs) or extended Kalman filters (EKFs) and an interacting multiple model (IMM) to estimate and predict the vehicle state. We propose to apply deep reinforcement learning (DRL) for the transmit power optimization. In particular, we consider an intersection and a lane-changing scenario and apply proximal policy optimization (PPO) and soft actor-critic (SAC) to train the DRL model. Testing results show satisfactory power control strategies confining estimate variances below given threshold. SAC achieves higher performance compared to PPO. [ABSTRACT FROM AUTHOR]

Motion planning is one of the most significant part in autonomous driving. Learning-based motion planning methods attract many researchers’ attention due to the abilities of learning from the environment and directly making decisions from the perception. The deep Q-network, as a popular reinforcement learning method, has achieved great progress in autonomous driving, but these methods seldom use the global path information to handle the issue of directional planning such as making a turning at an intersection since the agent usually learns driving strategies only by the designed reward function, which is difficult to adapt to the driving scenarios of urban roads. Moreover, different motion commands such as the steering wheel and accelerator are associated with each other from classic Q-networks, which easily leads to an unstable prediction of the motion commands since they are independently controlled in a practical driving system. In this paper, a conditional deep Q-network for directional planning is proposed and applied in end-to-end autonomous driving, where the global path is used to guide the vehicle to drive from the origination to the destination. To handle the dependency of different motion commands in Q-networks, we take use of the idea of fuzzy control and develop a defuzzification method to improve the stability of predicting the values of different motion commands. We conduct comprehensive experiments in the CARLA simulator and compare our method with the state-of-the-art methods. Experimental results demonstrate the proposed method achieves better learning performance and driving stability performance than other methods. [ABSTRACT FROM AUTHOR]

Vision-based deep learning (DL) methods have made great progress in learning autonomous driving models from large-scale crowd-sourced video datasets. They are trained to predict instantaneous driving behaviors from video data captured by on-vehicle cameras. In this paper, we develop a geo-context aware visualization system for the study of Autonomous Driving Model (ADM) predictions together with large-scale ADM video data. The visual study is seamlessly integrated with the geographical environment by combining DL model performance with geospatial visualization techniques. Model performance measures can be studied together with a set of geospatial attributes over map views. Users can also discover and compare prediction behaviors of multiple DL models in both city-wide and street-level analysis, together with road images and video contents. Therefore, the system provides a new visual exploration platform for DL model designers in autonomous driving. Use cases and domain expert evaluation show the utility and effectiveness of the visualization system. [ABSTRACT FROM AUTHOR]

Latest technological improvements increased the quality of transportation. New data-driven approaches bring out a new research direction for all control-based systems, e.g., in transportation, robotics, IoT and power systems. Combining data-driven applications with transportation systems plays a key role in recent transportation applications. In this paper, the latest deep reinforcement learning (RL) based traffic control applications are surveyed. Specifically, traffic signal control (TSC) applications based on (deep) RL, which have been studied extensively in the literature, are discussed in detail. Different problem formulations, RL parameters, and simulation environments for TSC are discussed comprehensively. In the literature, there are also several autonomous driving applications studied with deep RL models. Our survey extensively summarizes existing works in this field by categorizing them with respect to application types, control models and studied algorithms. In the end, we discuss the challenges and open questions regarding deep RL-based transportation applications. [ABSTRACT FROM AUTHOR]

In autonomous driving, many learning-based methods for motion planing have been proposed in literature, which can predict motion commands directly from the sensory data of the environment, but these methods can neither predict multiple motion commands, such as steering angle, accelerator and brake, nor balance errors among different motion commands. In this paper, we propose a deep cascaded neural network for predicting multiple motion commands which can be trained in an end-to-end manner for autonomous driving. The proposed deep cascaded neural network consists of a convolutional neural network (CNN) and three long short-term memory (LSTM) units, fed with images from a front-facing camera installed at the vehicle. As the outputs, the proposed model can predict thee motion planning commands simultaneously including steering angle, acceleration, and brake to enable the autonomous driving. In order to balance errors among different motion commands and improve prediction accuracy, we propose a new network training algorithm, where three independent loss functions are designed to separately update the weights in the three LSTMs connected to three motion commands. We conduct comprehensive experiments using the data from a driving simulator and compare our method with the state-of-the-art methods. Simulation results demonstrate the proposed motion planning model achieves better accuracy performance than other models. [ABSTRACT FROM AUTHOR]

The localization system is one of the key components of autonomous vehicles. Widely-used GNSS (low-cost) alone can not meet the centimeter accuracy and is unstable in urban environment. Typically LiDAR-based localization system is very computationally demanding due to the usage of the dense point cloud map. In this paper, we propose a novel lightweight LiDAR-based localization system for autonomous vehicle in this paper. The proposed system only relies on lightweight poles and curbs landmark map, which is highly robust and efficient compared to other localization systems. Poles and curbs are selected as landmarks because of their commonality and stability. We novelly propose a Branch-and-Bound (BnB)-based global optimization method to tackle the data association problem of poles. Motion decoupling is adopted to decouple translation and rotation to improve the efficiency of the BnB-based algorithm. Besides, we propose a new local grid map-based representation for curbs to make better use of curb information. Cost functions for pole and curb are defined respectively and then fused for the subsequent non-linear optimization method to obtain the vehicle location. Experiments on KITTI dataset and our self-collected dataset demonstrate the efficiency and accuracy of our system. [ABSTRACT FROM AUTHOR]

By considering a lane change maneuver as primarily a longitudinal motion planning problem, this paper presents a lane change maneuver algorithm with a pragmatic approach to determine an inter-vehicle traffic gap and time instance to perform the maneuver. The proposed approach selects an appropriate inter-vehicle traffic gap and time instance to perform the lane change maneuver by simply estimating whether there might exist a longitudinal trajectory that allows the automated vehicle to safely perform the maneuver. The lane change maneuver algorithm then proceeds to solve two loosely coupled convex quadratic programs to obtain the longitudinal trajectory to position the automated vehicle in the selected inter-vehicle traffic gap at the desired time instance and the corresponding lateral trajectory. Simulation results demonstrate the capability of the proposed approach to select an appropriate inter-vehicle traffic gap and time instance to initialize the lateral motion of a lane change maneuver in various traffic scenarios. The real-time ability of the lane change maneuver algorithm to generate safe and smooth trajectories is shown by experimental results of a Volvo V60 performing automated lane change maneuvers on a test track. [ABSTRACT FROM PUBLISHER]

3D object detection from LiDAR point cloud is a challenging problem in 3D scene understanding and has many practical applications. In this paper, we extend our preliminary work PointRCNN to a novel and strong point-cloud-based 3D object detection framework, the part-aware and aggregation neural network (Part- $A^2$ A 2 net). The whole framework consists of the part-aware stage and the part-aggregation stage. First, the part-aware stage for the first time fully utilizes free-of-charge part supervisions derived from 3D ground-truth boxes to simultaneously predict high quality 3D proposals and accurate intra-object part locations. The predicted intra-object part locations within the same proposal are grouped by our new-designed RoI-aware point cloud pooling module, which results in an effective representation to encode the geometry-specific features of each 3D proposal. Then the part-aggregation stage learns to re-score the box and refine the box location by exploring the spatial relationship of the pooled intra-object part locations. Extensive experiments are conducted to demonstrate the performance improvements from each component of our proposed framework. Our Part- $A^2$ A 2 net outperforms all existing 3D detection methods and achieves new state-of-the-art on KITTI 3D object detection dataset by utilizing only the LiDAR point cloud data. [ABSTRACT FROM AUTHOR]

With a goal to improve transportation safety, this paper proposes a collaborative driving framework based on assessments of both internal and external risks involved in vehicle driving. The internal risk analysis includes driver drowsiness detection and driver intention recognition that helps to understand the human driver’s behavior. Steering wheel data and facial expression are used to detect the driver’s drowsiness. Hidden Markov models are adapted to recognize the driver’s intention using the vehicle’s lane position, control, and state data. For the external risk analysis, a co-pilot utilizes a collision avoidance system to estimate the collision probability between the ego vehicle and other nearby vehicles. Based on the risk analyses, we design a novel collaborative driving scheme by fusing the control inputs from the human driver and the co-pilot to obtain the final control input for the ego vehicle under different circumstances. The proposed collaborative driving framework is validated in an assisted-driving testbed, which enables both autonomous and manual driving capabilities. [ABSTRACT FROM AUTHOR]

Semantic segmentation (SS) partitions an image into several coherent semantically meaningful parts and classifies each part into one of the pre-determined classes. In this paper, we argue that the existing SS methods cannot be reliably applied to autonomous driving system as they ignore the different importance levels of distinct classes for safe driving. For example, pedestrian, car, and bicyclist in the scene are much more important than sky and building when driving a car, so their segmentations should be as accurate as possible. To incorporate the importance information possessed by various object classes, this paper designs an “importance-aware loss” (IAL) that specifically emphasizes the critical objects for autonomous driving. The IAL operates under a hierarchical structure and the classes with different importance are located in different levels so that they are assigned distinct weights. Furthermore, we derive the forward and backward propagation rules for IAL and apply them to four typical deep neural networks for realizing SS in an intelligent driving system. The experiments on CamVid and Cityscapes data sets reveal that, by employing the proposed loss function, the existing deep learning models, including FCN, SegNet, ENet, and ERFNet, are able to consistently obtain the improved segmentation results on the pre-defined important classes for safe driving. [ABSTRACT FROM AUTHOR]

Lane detection on road segments with complex topologies such as lane merge/split and highway ramps is not yet a solved problem. This paper presents a novel graph-embedded solution. It consists of two key parts, a learning-based low-level lane feature extraction algorithm, and a graph-embedded lane inference algorithm. The former reduces the over-reliance on customized annotated/labeled lane data. We leveraged several open-source semantic segmentation datasets (e.g., Cityscape, Vistas, and Apollo) and designed a dedicated network that can be trained across these heterogeneous datasets to extract lane attributes. The latter algorithm constructs a graph to represent the lane geometry and topology. It does not rely on strong geometric assumptions such as lane lines are a set of parallel polynomials. Instead, it constructs a graph based on detected lane nodes. The lane parameters in the world coordinate are inferred by efficient graph-based searching and calculation. The performance of the proposed method is verified on both open source and our own collected data. On-vehicle experiments were also conducted and the comparison with Mobileye EyeQ2 shows favorable results. [ABSTRACT FROM AUTHOR]

This paper presents a 64×64 dual mode flash LIDAR sensor that utilizes a triple integration timing technique. The sensor, fabricated in 130 nm HV CMOS, is capable of operating in both a direct time-of-flight (ToF) mode where the timestamp of the first arriving photon is recorded per-pixel, as well as a photon counting mode where the number of photons is recorded over a time interval. The timing technique utilizes both a time-to-digital converter (TDC) and a time-to-amplitude converter (TAC) with a counter measuring global clock cycles and the triple integration interpolator (TII) measuring between clock cycles. The TII uses an analog integration with an additional two reference integrations allowing the time measurement to be resistant to PVT variation and in turn, allowing the circuit to be miniaturized without causing a large timing non-uniformity across the array. Utilizing the TII, the sensor achieves a state-of-the-art timing performance, with a resolution of 13 ps (1.95 mm depth resolution), a maximum range of 220 μs (32 km), a single-shot jitter of 233 ps, and a differential non-linearity (DNL) of 6 ps (0.47 LSB). The sensor captures at a maximum frame rate of 8,300 fps and consumes 733 mW during operation. Experimental scenarios demonstrating the operation of the sensor are also provided. [ABSTRACT FROM AUTHOR]

3D neural networks are widely used in real-world applications (e.g., AR/VR headsets, self-driving cars). They are required to be fast and accurate; however, limited hardware resources on edge devices make these requirements rather challenging. Previous work processes 3D data using either voxel-based or point-based neural networks, but both types of 3D models are not hardware-efficient due to the large memory footprint and random memory access. In this paper, we study 3D deep learning from the efficiency perspective. We first systematically analyze the bottlenecks of previous 3D methods. We then combine the best from point-based and voxel-based models together and propose a novel hardware-efficient 3D primitive, Point-Voxel Convolution (PVConv). We further enhance this primitive with the sparse convolution to make it more effective in processing large (outdoor) scenes. Based on our designed 3D primitive, we introduce 3D Neural Architecture Search (3D-NAS) to explore the best 3D network architecture given a resource constraint. We evaluate our proposed method on six representative benchmark datasets, achieving state-of-the-art performance with 1.8-23.7× measured speedup. Furthermore, our method has been deployed to the autonomous racing vehicle of MIT Driverless, achieving larger detection range, higher accuracy and lower latency. [ABSTRACT FROM AUTHOR]

Adaptive control theory has ushered in a fruitful era for the research and development of intelligent ground vehicle transportation systems. Notably, owing to its intelligence of handling parametric uncertainties via online learning and adaptation, the adaptive control methodology has attracted a great deal of attention in tackling autonomous/automated vehicle control problems. In this paper, we aim to improve the existing adaptive-control-based path-following controllers from two aspects. First, a non-quadratic-Lyapunov-function-based model reference adaptive controller is synthesized to achieve enhanced $\mathcal {L}^{\mathbf {1+\alpha }}$ tracking performance. Second, a $\mathcal{C}^{ \boldsymbol {\infty }}$ -differentiable smooth parameter projection scheme is employed for preventing the disturbance-induced control parameter drift. The stability of the redesigned path-tracking adaptive controller is analyzed. Furthermore, validations and comparative studies are conducted via hardware-in-the-loop experiments. [ABSTRACT FROM AUTHOR]

Game engines are increasingly used as simulation platforms by the autonomous vehicle community to develop vehicle control systems and test environments. A key requirement for simulation-based development and verification is determinism, since a deterministic process will always produce the same output given the same initial conditions and event history. Thus, in a deterministic simulation environment, tests are rendered repeatable and yield simulation results that are trustworthy and straightforward to debug. However, game engines are seldom deterministic. This paper reviews and identifies the potential causes and effects of non-deterministic behaviours in game engines. A case study using CARLA, an open-source autonomous driving simulation environment powered by Unreal Engine, is presented to highlight its inherent shortcomings in providing sufficient precision in experimental results. Different configurations and utilisations of the software and hardware are explored to determine an operational domain where the simulation precision is sufficiently high i.e. variance between repeated executions becomes negligible for development and testing work. Finally, a method of a general nature is proposed, that can be used to find the domains of permissible variance in game engine simulations for any given system configuration. [ABSTRACT FROM AUTHOR]

In recent years, considerable progress has been made in semantic segmentation of images with favorable environments. However, the environmental perception of autonomous driving under adverse weather conditions is still very challenging. In particular, the low visibility at nighttime greatly affects driving safety. In this paper, we aim to explore image segmentation in low-light scenarios, thereby expanding the application range of autonomous vehicles. The segmentation algorithms for road scenes based on deep learning are highly dependent on the volume of images with pixel-level annotations. Considering the scarcity of labeled large-scale nighttime data, we performed synthetic data collection and data style transfer using images acquired in daytime based on the autonomous driving simulation platform and generative adversarial network, respectively. In addition, we also proposed a novel nighttime segmentation framework (SFNET-N) to effectively recognize objects in dark environments, aiming at the boundary blurring caused by low semantic contrast in low-illumination images. Specifically, the framework comprises a light enhancement network which introduces semantic information for the first time and a segmentation network with strong feature extraction capability. Extensive experiments with Dark Zurich-test and Nighttime Driving-test datasets show the effectiveness of our method compared with existing state-of-the art approaches, with 56.9% and 57.4% mIoU (mean of category-wise intersection-over-union) respectively. Finally, we also performed real-vehicle verification of the proposed models in road scenes of Zhenjiang city with poor lighting. The datasets are available at https://github.com/pupu-chenyanyan/semantic-segmentation-on-nightime. [ABSTRACT FROM AUTHOR]

In this paper, a three-level decision-making framework is developed to generate safe and effective decisions for autonomous vehicles (AVs). A key component in this decision framework is a normal-form game to capture the interactions between the ego vehicle and its surrounding vehicles. The payoffs in the normal-form game are designed to capture both safety reward and the reward gained by obeying (or the price paid by violating) “soft” traffic rules, e.g., first-come-first-go. This game formulation enables the ego to 1) make appropriate decisions considering the payoffs and possible actions of its surrounding vehicles, and 2) take intelligent actions in emergencies that may sacrifice some soft traffic rules to ensure safety. Moreover, we introduce parameters in the payoff matrix to tune the ego vehicle’s behavior, e.g., aggressiveness level. A neural network is developed to learn the tuning parameters via supervised learning. In addition, to enable the ego to respond timely to different surrounding vehicles’ driving styles, driving style characterization is incorporated into the payoff design for the normal-form game. Simulation studies are conducted to demonstrate the performance of the developed algorithms in two-vehicle intersection-crossing and lane-changing scenarios. [ABSTRACT FROM AUTHOR]

For the autonomous driving system, accurately recognizing the actions of different roles in the traffic scene is the prerequisite for realizing this kind of human-vehicle information interaction. In this paper, we propose a complete framework based on 3D human pose estimation to recognize the actions of different roles on the road. The main objects recognized include traffic police, cyclists, and some passersby in need. We perform action recognition based on a dynamic adaptive graph convolutional network, which can realize the action recognition of objects based on 3D human pose. In addition to the action recognition module, we have optimized both the object detection module and the human pose estimation module in the framework so that the framework can handle multiple objects at the same time, which can be closer to the real traffic scene. To realize complex and changeable human action recognition, we built a multi-view camera system to collect responsible 3D human pose datasets containing traffic police gestures, cyclist gestures, and pedestrians’ body movements. In the experiments, compared to other state-of-the-art researches, the proposed framework can achieve comparable results with the same dataset. Satisfactory performance has also been obtained on the real data we collected, which can handle a variety of different action recognition tasks at the same time. [ABSTRACT FROM AUTHOR]

Automated driving is predicted to enhance traffic safety, transport efficiency, and driver comfort. To extend the capability of current advanced driver assistance systems, and eventually realize fully automated driving, the intelligent vehicle system must have the ability to plan different maneuvers while adapting to the surrounding traffic environment. This paper presents an algorithm for longitudinal and lateral trajectory planning for automated driving maneuvers where the vehicle does not have right of way, i.e., yielding maneuvers. Such maneuvers include, e.g., lane change, roundabout entry, and intersection crossing. In the proposed approach, the traffic environment which the vehicle must traverse is incorporated as constraints on its longitudinal and lateral positions. The trajectory planning problem can thereby be formulated as two loosely coupled low-complexity model predictive control problems for longitudinal and lateral motion. Simulation results demonstrate the ability of the proposed trajectory planning algorithm to generate smooth collision-free maneuvers which are appropriate for various traffic situations. [ABSTRACT FROM PUBLISHER]

Laser scanners are widely used as the primary sensor for autonomous driving. When the commercialization of autonomous driving is considered, a 2.5-D multi-layer laser scanner is one of the best sensor options. In this paper, a new method is presented to detect pedestrians and vehicles using a 2.5-D multi-layer laser scanner. The proposed method consists of three steps: 1) segmentation; 2) feature extraction; and 3) classification; this paper focuses on the last two steps. In feature extraction, new features for the multi-layer laser scanner are proposed to improve the classification performance. In classification, radial basis function additive kernel support vector machine is employed to reduce the computation time while maintaining the performance. The proposed method is implemented on a real vehicle, and its performance is tested in a real-world environment. The experiments indicate that the proposed method has good performance in many real-life situations. [ABSTRACT FROM PUBLISHER]

In the context of vehicle trajectory planning, motion primitives are trajectories connecting pairs of boundary conditions. In autonomous racing, motion primitives have been used as computationally faster alternatives to model predictive control, for online obstacle avoidance. However, the existing motion primitive formulations are either simplified and suboptimal, or computationally expensive for accurate collision avoidance. This paper introduces new motion primitives for autonomous racing, aiming to accurately approximate the minimum-time vehicle trajectories while ensuring computational efficiency. We present a novel neural network, named PathPoly-NN, whose internal architecture is designed to learn the minimum-time vehicle path. Our motion primitives combine PathPoly-NN with a fast forward-backward method to compute the minimum-time speed profile. Compared to existing neural networks, PathPoly-NN generalizes better with small training sets, and it has better accuracy in approximating the minimum-time path. Additionally, our motion primitives have lower computational burden and higher accuracy than existing methods based on cubic polynomials and $G^{2}$ clothoid curves. Finally, the motion primitives of this paper achieve similar maneuver times as minimum-time economic nonlinear model predictive control (E-NMPC), but with significantly lower computational load (two orders of magnitude). The results open promising perspectives of applications in graph-based trajectory planners for autonomous racing.

Lane detection is a key technical challenge in autonomous driving and advanced driving assistance systems. The current lane detection system has insufficient robustness and low detection accuracy problem in complex and harsh environmental conditions, and lacks research on visual cues without information. In order to accurately and effectively detect multiple lanes in challenging scenarios and meet the real-time requirements of actual driving, a lane detection method based on recurrent cross attention mechanism is proposed in this paper. Firstly, by embedding the recurrent cross attention module in the feature extraction module, the contextual information of surrounding pixels on the intersection path is obtained, and useful information is collected from other regions for accurate lane detection; Secondly, anchor point detection is selected based on the prior position of the designed lane starting position; Finally, combining mixed anchor points to represent lane coordinates, using row and column anchors to learn lane coordinates and improve the real-time performance of lane detection models. The detection model RCCLaneNet designed in this paper was evaluated on the TuSimple and CULane datasets. The experimental results showed that the F1 of this method can reach 97.05% on the TuSimple dataset, 78.5% on the CULane dataset, and the real-time inference speed index FPS can reach +150. The proposed model achieves advanced performance in lane detection tasks in complex scenes while maintaining real-time detection efficiency, and has good universality in actual driving scenarios.

Road target detection is essential for enhancing vehicle safety, increasing operational efficiency, and optimizing user experience. It also forms a crucial part of autonomous driving and intelligent monitoring systems. However, current technologies face significant limitations in multi-level feature fusion and the accurate identification of key targets in complex data environments. To address these challenges, this paper proposes an innovative algorithmic model called BiFPN GAM SimC2f-YOLO (BGS-YOLO), aimed at improving detection performance. Initially, this paper employs the Bidirectional Feature Pyramid Network (BiFPN) to effectively integrate multi-level features. This integration overcomes the limitations in feature extraction and recognition found in existing target detection algorithms. Following this, this paper introduces the Global Attention Module (GAM), which markedly improves the efficiency and accuracy of extracting key target information in complex data environments. Additionally, this paper innovatively designs the SimAM-C2f (SimC2f) network, further advancing feature expressiveness and fusion efficiency. Experiments on the public COCO dataset demonstrate that the BGS-YOLO model significantly outperforms the existing YOLOv8n model. Notably, it shows a 7.3% increase in mean average precision (mAP) and a 2.4% improvement in accuracy. These results highlight the model’s high precision and swift response in detecting road targets in complex traffic scenarios. Consequently, the BGS-YOLO model has the potential to significantly enhance road safety and contribute to a considerable reduction in traffic accident rates.

A safe trajectory planning for on-road autonomous driving is a challenging problem owing to the variety and complexity of driving environments. The problem should involve the consideration of numerous aspects such as road geometry, lane-structured roads, traffic regulations, traffic participants, and vehicle physical limitations. Furthermore, dynamically changing movements of surrounding vehicles make the problem more challenging. It requires the planner’s ability to react to the changes in the driving environments in real time. To solve this problem, sampling and numerical optimization-based trajectory planners were introduced. However, these methods have their own limitations in generating a safe trajectory in these dynamic scenarios. To overcome these issues, this paper proposes a hybrid trajectory planning scheme to integrate the strength of the sampling and optimization methods. With the sampling method for a lateral movement, the planner can deal with various trajectories with multiple maneuvers. This helps the planner to generate a reactive trajectory in a dynamically changing environment. The numerical optimization of a longitudinal movement enables the planner to adapt to diverse situations without restriction of predefined patterns for specific driving purposes. The proposed method was implemented with an embedded optimization coder and C++ environment. Based on this, its performance was evaluated through simulation and real driving tests in various on-road dynamic scenarios. [ABSTRACT FROM AUTHOR]

The Internet of Vehicles (IoV) as a promising application of Internet of Things (IoT) has played a significant role in autonomous driving, by connecting intelligent vehicles. Autonomous driving needs to process the mass environmental sensing data in coordination with surrounding vehicles, and makes an accurate driving judgment accordingly. Since the vehicles always have limited computing resources, processing these data in parallel with efficient task scheduling is one of the most important topics. Most current work focuses on formulating special scenarios and service requirements as optimization problems. However, the complicated and dynamic environment of vehicular computing is hard to model, predict and control, making those previous methods unscalable and unable to reflect the real scenario. In this paper, a Multi-task Deep reinforcement learning approach for scalable parallel Task Scheduling (MDTS) is firstly devised. For avoiding the curse of dimensionality when coping with complex parallel computing environments and jobs with diverse properties, we extend the action selection in Deep Reinforcement Learning (DRL) to a multi-task decision, where the output branches of multi- task learning are fine-matched to parallel scheduling tasks. Child tasks of a job are accordingly assigned to distributed nodes without any human knowledge while the resource competition among parallel tasks is leveraged through shared neural network layers. Moreover, we design an appropriate reward function to optimize multiple metrics simultaneously, with emphasis on specific scenarios. Extensive experiments show that the MDTS significantly increases the overall reward compared with least- connection scheduling and particle swarm optimization algorithm from −16.71, −0.67 to 2.93, respectively. [ABSTRACT FROM AUTHOR]

Vehicle-to-vehicle communication is a key technology for achieving increased perception for automated vehicles, where the communication enables virtual sensing by means of sensors in other vehicles. In addition, this technology also allows detection and recognition of objects that are out-of-sight. This paper presents a trust system that allows a cooperative and automated vehicle to make more reliable and safe 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 surrounding vehicles. This research goes beyond secure communication and concerns the verification of the received data on a system level. The results show that the proposed method is capable of correctly identifying various traffic situations and how the trust index is used while manoeuvring in a platoon merge scenario. [ABSTRACT FROM PUBLISHER]