Search

Estimating depth from RGB images is a long-standing ill-posed problem, which has been explored for decades by the computer vision, graphics, and machine learning communities. Among the existing techniques, stereo matching remains one of the most widely used in the literature due to its strong connection to the human binocular system. Traditionally, stereo-based depth estimation has been addressed through matching hand-crafted features across multiple images. Despite the extensive amount of research, these traditional techniques still suffer in the presence of highly textured areas, large uniform regions, and occlusions. Motivated by their growing success in solving various 2D and 3D vision problems, deep learning for stereo-based depth estimation has attracted a growing interest from the community, with more than 150 papers published in this area between 2014 and 2019. This new generation of methods has demonstrated a significant leap in performance, enabling applications such as autonomous driving and augmented reality. In this paper, we provide a comprehensive survey of this new and continuously growing field of research, summarize the most commonly used pipelines, and discuss their benefits and limitations. In retrospect of what has been achieved so far, we also conjecture what the future may hold for deep learning-based stereo for depth estimation research. [ABSTRACT FROM AUTHOR]

This paper aims to resolve a clear and unrealistic limitation found in recent works that have proposed solutions to the massive demand for the autonomous electric mobility-on-demand (AEMoD) services. Namely, this paper will focus on the matched charge-to-trip only service by enabling sub-class service; i.e., allowing vehicles to serve customer classes with trips needing less charge. The paper proposes a multi-class management system with a new queuing model for one city zone, and then derives the stability conditions of the system. Proportions of vehicles from each class that will immediately serve customers with sub-class service or full/partial charge are modeled, and the maximum expected response time of the system is then minimized by optimizing these vehicle service decisions. The advantages of the new model are then compared with previously proposed schemes and non-optimized models through simulations. [ABSTRACT FROM AUTHOR]

In this paper, we develop a novel visual-inertial navigation system with motion constraint (VINS-Motion), which extends the visual-inertial navigation system (VINS) to incorporate vehicle motion information for improving the autonomous vehicles localization accuracy. We introduce the vehicle stop detection module in the pre-processing measurement procedure. During the back-end processing, if the vehicle is detected to be in a stopping state, we perform optimization based on the stop constraint to help eliminate the abnormal jitter of the estimated pose, thus ensuring the reasonability of the trajectory. Otherwise, besides the prior information, IMU measurement residual, and visual measurement residual utilized in VINS, vehicle orientation/velocity constraint built by Ackerman steering model is first exploited to constitute residuals. We minimize the sum of priors and Mahalanobis norms of three kinds of residuals to obtain a maximum posterior estimation, thus increasing system consistency and accuracy. The proposed approach is validated on public datasets and compared with state-of-the-art algorithms, which demonstrates that the motion constraint proposed in this paper improves the standard VINS performance and achieves significantly higher positioning accuracy. [ABSTRACT FROM AUTHOR]

Highway platooning of vehicles has been identified as a promising framework in developing intelligent transportation systems. By autonomous or semi-autonomous vehicle control and intervehicle coordination, an appropriately managed platoon can potentially offer enhanced safety, improved highway utility, increased fuel economy, and reduced emissions. This paper is focused on quantitative characterization of the impact of communication information structures and contents on platoon safety. By comparing different information structures that combine front sensors, rear sensors, and wireless communication channels, as well as different information contents such as distances, speeds, and drivers' actions, we reveal a number of intrinsic relationships between vehicle coordination and communications in platoons. Typical communication standards and related communication latency are used as benchmark cases in this paper. The findings of this paper provide useful guidelines in sensor selections, communication resource allocations, and vehicle coordination. [ABSTRACT FROM AUTHOR]

Efficient point cloud compression is essential for applications like virtual and mixed reality, autonomous driving, and cultural heritage. This paper proposes a deep learning-based inter-frame encoding scheme for dynamic point cloud geometry compression. We propose a lossy geometry compression scheme that predicts the latent representation of the current frame using the previous frame by employing a novel feature space inter-prediction network. The proposed network utilizes sparse convolutions with hierarchical multiscale 3D feature learning to encode the current frame using the previous frame. The proposed method introduces a novel predictor network for motion compensation in the feature domain to map the latent representation of the previous frame to the coordinates of the current frame to predict the current frame’s feature embedding. The framework transmits the residual of the predicted features and the actual features by compressing them using a learned probabilistic factorized entropy model. At the receiver, the decoder hierarchically reconstructs the current frame by progressively rescaling the feature embedding. The proposed framework is compared to the state-of-the-art Video-based Point Cloud Compression (V-PCC) and Geometry-based Point Cloud Compression (G-PCC) schemes standardized by the Moving Picture Experts Group (MPEG). The proposed method achieves more than 88% BD-Rate (Bjøntegaard Delta Rate) reduction against G-PCCv20 Octree, more than 56% BD-Rate savings against G-PCCv20 Trisoup, more than 62% BD-Rate reduction against V-PCC intra-frame encoding mode, and more than 52% BD-Rate savings against V-PCC P-frame-based inter-frame encoding mode using HEVC. These significant performance gains are cross-checked and verified in the MPEG working group. [ABSTRACT FROM AUTHOR]

Uncertain cut-in maneuver of vehicles from adjacent lanes makes it difficult for vehicle's automatic speed control strategy to make judgments and effective control decisions. In this paper, an intelligent speed control strategy for uncertain cut-in scenarios is established based on a basic autonomous driving system. This strategy judges cut-in maneuver from surrounding vehicles and outputs adaptive control action under current environment according to Q value of state-action pair based on a Q network. In addition, according to the analysis of cut-in scenarios, the Q network is trained based on a novel reinforcement learning method named as experience screening deep Q-learning network (ES-DQN). The proposed ES-DQN is an extension of double deep Q-learning network (DDQN) algorithm, and includes two parts: experience screening and policy learning. Based on the experience screened from the experience screening part, the proposed learning method can train an intelligent speed control strategy which has stronger adaptability and control effect in uncertain cut-in scenarios. According to simulation results, the proposed intelligent speed control strategy trained by ES-DQN has better performance under uncertain cut-in scenarios than DDQN method and traditional ACC strategy. Meanwhile, by adjusting weight value in reward function, the system can realize different control target. [ABSTRACT FROM AUTHOR]

Recognizing wet surfaces and their degrees of wetness is essential for many computer vision applications. Surface wetness can inform us slippery spots on a road to autonomous vehicles, muddy areas of a trail to humanoid robots, and the freshness of groceries to us. The fact that surfaces darken when wet, i.e., monochromatic appearance change, has been modeled to recognize wet surfaces in the past. In this paper, we show that color change, particularly in its spectral behavior, carries rich information about surface wetness. We first derive an analytical spectral appearance model of wet surfaces that expresses the characteristic spectral sharpening due to multiple scattering and absorption in the surface. We present a novel method for estimating key parameters of this spectral appearance model, which enables the recovery of the original surface color and the degree of wetness from a single multispectral image. Applied to a multispectral image, the method estimates the spatial map of wetness together with the dry spectral distribution of the surface. To our knowledge, this is the first work to model and leverage the spectral characteristics of wet surfaces to decipher its appearance. We conduct comprehensive experimental validation with a number of wet real surfaces. The results demonstrate the accuracy of our model and the effectiveness of our method for surface wetness and color estimation. [ABSTRACT FROM AUTHOR]

Many autonomous vehicles and advanced driver-assistance systems are equipped with front-facing cameras that detect and track objects using deep-learning-based algorithms. However, the localization capability of monocular cameras is often overlooked. In this paper, a novel method for estimating the pixel-wise error in a detected object's location versus its ground truth is proposed. As the object moves away from the camera, the pixel errors are shown to be normally distributed with unique spreads along the image's vertical axis (y-pixel). The pixel error appears to be smaller as objects get farther away, while at the same distance range, objects have similar error distribution across the camera's horizontal view. The horizontal axis (x-pixel) error appears to be smaller while the distance moves further away. However, the x-pixel location along a constant y-pixel row has no impact on the error distribution. The estimated x and y-pixel error distributions can in turn be used to form a spatial error distribution for finding the location of a detected object within a certain confidence interval. The spatial errors are then projected onto the world coordinate system using a camera transformation matrix to give a more realistic sense of what this error means. The results show that location estimation using monocular cameras generates an elliptical error distribution around the object with a larger error in the y-pixel direction compared to the x-pixel direction. This error distribution can be important to fuse information from multiple range-detecting sensors as well as multi-vehicle and multi-object tracking. The uncertainty characterization for position measurement, as demonstrated in this paper is an essential element of tracking and, is sensor and algorithm dependent. [ABSTRACT FROM AUTHOR]

The air-ground cooperative emergency networks can assist with the rapid reconstruction of communication in the disaster area, where unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) are deployed as base stations. The trajectory optimization of emergency base stations is of vital importance to the communication performance, which is related to the timeliness and effectiveness of rescue. In this paper, federated multi-agent deep deterministic policy gradient (F-MADDPG) based trajectory optimization algorithm is proposed to maximize the average spectrum efficiency. Specifically, the property of MADDPG is inherited to jointly control of multiple vehicles and federated averaging (FA) is utilized to eliminate the isolation of data to accelerate the convergence. Distributed F-MADDPG (DF-MADDPG) is further designed to reduce the communication overhead with a distributed architecture. The simulation results indicate that the proposed F-MADDPG and DF-MADDPG based algorithms significantly outperform the existing trajectory optimization algorithms, in terms of the average spectrum efficiency and the speed of convergence. [ABSTRACT FROM AUTHOR]

Contemporary and emergent automotive systems are heavily populated by complex integrated electronics. The number of safety-critical devices used in advanced driver-assistance systems or autonomous vehicles is growing with high-end models containing hundreds of embedded microcontrollers. Achieving functionally safe automotive electronics requires test solutions that might be costly to engineer. Therefore, to address challenges posed by high-quality and long-term reliability requirements, this article presents low-cost test pattern generation schemes for a scan-based hybrid logic BIST of automotive ICs. It may allow one to optimize test coverage and test time during in-system test applications. The first presented technique deploys a seed-flipping PRPG to periodically complement PRPG stages in a methodical tree-traversal manner. The second scheme is based on a seed-sorting approach that allows additional tradeoffs between test data volume and test coverage. As shown in this article, the proposed schemes can be easily integrated with a test compression environment and deployed in different modes of in-system testing, such as key-off, key-on, and periodic (incremental) online tests. Experimental results obtained for automotive designs and reported herein show improvements in test quality over conventional logic BIST schemes. [ABSTRACT FROM AUTHOR]

In order to improve the trajectory tracking accuracy and vehicle lateral stability, the paper proposes a lateral and longitudinal dynamics control framework of autonomous vehicles considering multi-parameter joint estimation. First, the multi-parameter joint estimation based on adaptive unscented Kalman filter (AUKF) is constructed to decouple and estimate the longitudinal position of vehicle's center of gravity (CG), tire-road friction coefficient (TRFC), tire cornering stiffness (TCS), tire vertical force (TVF) and road grade. Then, focusing on the large lateral acceleration condition, a lateral control based on the optimal front-tire lateral force is proposed by constructing the linear quadratic regulator (LQR) and combining the multi-parameter joint estimation. Additionally, a longitudinal control based on the drive and brake force compensation to realize the accurate speed tracking by combining road slop estimation is achieved. The fuzzy system employs the drive and brake force deviation as the input to implement the compensation of the throttle and brake. The proposed estimation and control framework are verified by co-simulation on PreScan and CarSim preliminarily. In order to further verify the effectiveness and practicability of algorithm, experiments are implemented on an autonomous vehicle platform, a hybrid Lincoln MKZ. Simulation and experimental results show that the proposed estimation and control framework possess excellent performance and enhance the tracking accuracy and lateral stability. [ABSTRACT FROM AUTHOR]

Vehicle-to-everything communication (V2X) and mobile edge computing (MEC) have introduced significant technological innovations for autonomous ground vehicles (AGVs), enabling most driving task computations to be performed with edge servers with higher quality and effectively reducing computational power consumption for AGVs. For reducing more computational power consumption which can reach hundreds of watts, sending more offloading tasks to edge servers can result in high latency and congestion when the network resources are finite and even cause serious traffic accidents. Besides, in V2X network, multiple transmission paths, including vehicle-to-vehicle (V2V) links and vehicle-to-infrastructure (V2I) links, can provide more valuable resources for sending offloading tasks, but they increase the complexity of resource management. To improve the utilization efficiency of network resources and ensure low power consumption for AGVs, first, this paper analyzes mathematical models of bandwidth, delay and power consumption based on different computing modes and transmission paths. Second, we propose an allocation algorithm for network resource allocation, transmission paths and computing modes to minimize power consumption according to the states of the V2X-MEC network measured and predicted by the navigation and positioning system. Third, a traversal approach simplified by a road segmentation method is adopted to solve for the optimal parameters. Through simulations, we compare our solution against an allocation approach based on the shortest path method, minimum average delay method and game theory method to illustrate the effectiveness of our proposed method in reducing the power consumption and transmission delay of uplinks. [ABSTRACT FROM AUTHOR]

Unsignalized intersection cooperation of connected and automated vehicles (CAVs) is able to eliminate green time loss of signalized intersections and improve traffic efficiency. Most of the existing research on unsignalized intersection cooperation considers fixed lane direction, where only specific turning behavior of vehicles is allowed on each lane. Given that traffic volume and the proportion of vehicles with different turning expectation may change with time, fixed lane direction may lead to inefficiency at intersections. This paper proposes a multi-lane unsignalized intersection cooperation method that considers flexible lane direction. The two-dimensional distribution of vehicles is calculated and vehicles that are not in conflict are scheduled to pass the intersection simultaneously. The formation reconfiguration method is utilized to achieve collision-free longitudinal and lateral position adjustment of vehicles. Simulations are conducted at different input traffic volumes and turning proportion of incoming vehicles, and the results indicate that our method outperformances the fixed-lane-direction unsignalized cooperation method and the signalized method. [ABSTRACT FROM AUTHOR]

Autonomous vehicles are expected to significantly enhance the human mobility. However, recently researchers have discovered and demonstrated some attacks on vehicles, which have caused a panic among the public. Furthermore, these attacks have demonstrated that the security issue is still one of the major challenges of vehicles. In this paper, we propose a novel edge computing based anomaly detection, coined edge computing based vehicle anomaly detection (EVAD), which exploits edge based sensor data fusion to identify the anomaly events. The time domain property, i.e., the correlation between different intra-vehicle sensors, and the frequency domain property of sensor data are utilized to judge whether an anomaly has occurred within the vehicle. Especially, to reduce the computation overhead and improve the performance, multiple sensors will be organized as ring architecture, which is a tradeoff of detection accuracy and complexity. In addition, the major components (e.g., anomaly detection module) of EVAD are embedded in edge computing devices, which make the anomaly detection be more efficient and privacy preserving. Meanwhile, a more appropriate model is generated on the cloud server, of which computation overhead maybe heavy for edge computing devices. This paper evaluates the performance of EVAD under different scenarios, and the experimental results demonstrate its feasibility and efficiency. The average true positive rate achieves 99.5% with 1% false positive rate. [ABSTRACT FROM AUTHOR]

Part I of this paper proposed a development process and a system platform for the development of autonomous cars based on a distributed system architecture. The proposed development methodology enabled the design and development of an autonomous car with benefits such as a reduction in computational complexity, fault-tolerant characteristics, and system modularity. In this paper (Part II), a case study of the proposed development methodology is addressed by showing the implementation process of an autonomous driving system. In order to describe the implementation process intuitively, core autonomous driving algorithms (localization, perception, planning, vehicle control, and system management) are briefly introduced and applied to the implementation of an autonomous driving system. We are able to examine the advantages of a distributed system architecture and the proposed development process by conducting a case study on the autonomous system implementation. The validity of the proposed methodology is proved through the autonomous car A1 that won the 2012 Autonomous Vehicle Competition in Korea with all missions completed. [ABSTRACT FROM PUBLISHER]

Visual localization enables autonomous vehicles to navigate in their surroundings and augmented reality applications to link virtual to real worlds. Practical visual localization approaches need to be robust to a wide variety of viewing conditions, including day-night changes, as well as weather and seasonal variations, while providing highly accurate six degree-of-freedom (6DOF) camera pose estimates. In this paper, we extend three publicly available datasets containing images captured under a wide variety of viewing conditions, but lacking camera pose information, with ground truth pose information, making evaluation of the impact of various factors on 6DOF camera pose estimation accuracy possible. We also discuss the performance of state-of-the-art localization approaches on these datasets. Additionally, we release around half of the poses for all conditions, and keep the remaining half private as a test set, in the hopes that this will stimulate research on long-term visual localization, learned local image features, and related research areas. Our datasets are available at visuallocalization.net , where we are also hosting a benchmarking server for automatic evaluation of results on the test set. The presented state-of-the-art results are to a large degree based on submissions to our server. [ABSTRACT FROM AUTHOR]

Transmissibility is a mathematical model that relates a subset of a system’s outputs to another subset of outputs of the same system without knowledge of the external excitation or the dynamics of the system. This study investigates fault detection, localization, and mitigation of connected autonomous vehicles (CAV) platoons using transmissibility operators. A CAV platoon is a network of connected autonomous vehicles that communicate together to move in a specific path with the desired velocity. Failure in a physical component of a vehicle, or failure in the form of an internal delay, a cyber-attack, or a communication time-delay affects the safety and security of the CAV platoons. In this paper, we use measurements from sensors available in CAV platoons to identify transmissibility operators, which are used for health monitoring, fault localization, and fault mitigation in the platoon. We first consider the case of vehicle-to-cloud communication (V2C) to monitor the platoon’s health. Then, we assume that the platoon loses communication with the cloud, and we monitor the health of the platoon based on vehicle-to-vehicle (V2V) communication. We apply the proposed technique to a model of the platoon obtained using the bond graph approach, and an experimental setup consisting of three connected autonomous robots. [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]

The problem of sensor fault estimation has been studied for Unmanned Surface Vehicle (USV) systems in networked environment. Assuming that the process noise is related to the measurement noise and packet loss will occur during the measurement transmission. We use an event-trigger mechanism (ETM) to save communication resources. Accordingly, an observer is proposed to estimate the state and sensor fault, and the trigger threshold is associated with the correlation coefficient. By adjusting the threshold, the observer performance and energy consumption can be balanced in an ideal situation. The observer designed in this paper reduces the effects of correlation noises effectively and good handling network packet losses, and it has been proved that the expected covariance of the estimation error is bounded. Finally, the simulation experiment on the identified model of USV in the laboratory is performed to validate the relevant conclusions. [ABSTRACT FROM AUTHOR]

This paper investigates the fundamentals of motion planning for minimizing motion sickness in transportation systems of higher automation levels. The optimum velocity profile is sought for a predefined road path from a specific starting point to a final one within specific and given boundaries and constraints in order to minimize the motion sickness and the journey time. An empirical approach based on British standard is used to evaluate motion sickness. The trade-off between minimizing motion sickness and journey time is investigated through multi-objective optimization by altering the weighting factors. The correlation between sickness and journey time is represented as a Pareto front because of their conflicting relation. The compromise between the two components is quantified along the curve, while the severity of the sickness is determined using frequency analysis. In addition, three case studies are developed to investigate the effect of driving style, vehicle speed, and road width, which can be considered among the main factors affecting motion sickness. According to the results, the driving style has higher impact on both motion sickness and journey time compared to the vehicle speed and the road width. The benefit of higher vehicle speed gives shorter journey time while maintaining relatively lower illness rating compared with lower vehicle speed. The effect of the road width is negligible on both sickness and journey time when travelling on a longer road. The results pave the path for the development of vehicular technologies to implement for real-world driving from the outcomes of this paper. [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]

This paper presents a novel motion planning and tracking framework for automated vehicles based on artificial potential field (APF) elaborated resistance approach. Motion planning is one of the key parts of autonomous driving, which plans a sequence of movement states to help vehicles drive safely, comfortably, economically, human-like, etc. In this paper, the APF method is used to assign different potential functions to different obstacles and road boundaries; while the drivable area is meshed and assigned resistance values in each edge based on the potential functions. A local current comparison method is employed to find a collision-free path. As opposed to a path, the vehicle motion or trajectory should be planned spatiotemporally. Therefore, the entire planning process is divided into two spaces, namely the virtual and actual. In the virtual space, the vehicle trajectory is predicted and executed step by step over a short horizon with the current vehicle speed. Then, the predicted trajectory is evaluated to decide if the speed should be kept or changed. Finally, it will be sent to the actual space, where an experimentally validated Carsim model controlled by a model predictive controller is used to track the planned trajectory. Several case studies are presented to demonstrate the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

Whether robot controllers are manually designed or synthesized from high-level task specifications, assumptions about the environment need to be made, which can involve adversarial events or cooperative robots. In either case, if these assumptions are violated at runtime, the robot will fail to fulfill its task and will likely do something unexpected. In this paper, we focus on controllers synthesized from linear temporal logic. We tackle the problem of making these controllers robust against environment assumption violations that are common in robot execution. Our solution is a three-layer system: first, we propose an offline approach that accounts for transient violations such that the robot can still complete its task after temporary anomalies; the second layer is an online approach that automatically relaxes the environment assumptions to better capture environment behaviors and that allows the robot to react accordingly; and, finally, we automatically modify the actual environment behaviors, when possible, through negotiation with one of the environment robots operating in the workspace, such that our assumptions are met by the other robot and both robots accomplish their tasks. [ABSTRACT FROM AUTHOR]

Fast and accurate visual scene understanding in autonomous vehicles is necessary but still very challenging. An autonomous vehicle must be taught to read the road like a human driver for better controlling the vehicle, so it is important to efficiently detect the road area and road markings. In this paper, we mainly focus on the vanishing point detection and its application in inverse perspective mapping (IPM) for road marking understanding. We first propose a fast and accurate vanishing point detection method for various types of roads, by adopting and improving Weber local descriptor to obtain salient representative texture and orientation information of the road area, and then voting for the dominant vanishing point with a simple line-voting scheme. Experimental results demonstrate that the proposed vanishing point detection approach gains a better performance than some state-of-the-art methods in terms of accuracy and computation time. Furthermore, we introduce the detected vanishing point into the IPM algorithm in the structured road environment, since some important calibration parameters can be automatically calculated by the vanishing point, especially on the rough road. Experiments also show that our proposed vanishing point-based IPM method is adaptive and accurate, which is conducive to the subsequent road marking detection and recognition. [ABSTRACT FROM AUTHOR]

To improve the structure layout flexibility and performance of the X-by-wire (XBW) ground vehicle, the control configured vehicle (CCV) principle is implemented on an XBW electric vehicle in this paper. By including the electric control system design together with the mechanical component design for the final layout, the CCV enhances the structure layout flexibility of the XBW ground vehicle considerably. Under the CCV principle, the closed-loop stability of the vehicle lateral dynamics system is improved by the involved yaw control system. In this paper, a robust pole assignment yaw controller with the consideration of parametric uncertainties is used. A full XBW electric vehicle testbed is shown as an example to demonstrate the benefit of the CCV of improving the hardware layout flexibility. Under the CCV, all the XBW components can be located flexibly to any position to enhance the configuration of the electric wires, saving of the space, and convenience of the assembling. The discussion of the yaw moment demand in different cases shows how the mechanical components and electric control systems influence and benefit each other in the CCV. The experiments based on the testbed show the desirable efficiency of the proposed yaw controller. [ABSTRACT FROM PUBLISHER]

This paper describes the modeling, analysis, and design of a complete (dc-to-dc) inductive wireless power transfer (WPT) system for industrial moving applications. The system operates at 6.78 MHz and delivers up to 150 W to a load moving along a linear path, providing a quasi-constant dc output voltage and maintaining a zero voltage switching operation, regardless of position and load, without any retuning or feedback. The inductive link consists of an array of stationary transmitting coils and a moving receiving coil whose length is optimized to achieve a constant coupling coefficient along the path. Each Tx coil is individually driven by a constant amplitude and phase sinusoidal current that is generated from a GaN-based coupled load-independent Class EF inverter. Two adjacent transmitters are activated at a given time depending on the receiver’s position; this effectively creates a virtual series connection between the two transmitting coils. The Rx coil is connected to a passive Class E rectifier that is designed to maintain a constant dc output voltage independent of its load and position. Extensive experimental results are presented to show the performance over different loading conditions and positions. A peak dc-to-dc efficiency of 80% is achieved at 100 W of dc output power and a dc output voltage variation of less than 5% is measured over a load range from 30 to $500~\Omega $ . The work in this paper is foreseen as a design solution for a high-efficient, maintenance-free, and reliable WPT system for powering sliders and mass movers in industrial automation plants. [ABSTRACT FROM PUBLISHER]

Path planning is a critical part for improving the driving safety and driver comfort of autonomous vehicles (AVs), especially in complex maneuvering conditions. In addition, different drivers have different preferences for AVs, thus, how to provide personalized trajectories for different drivers is a vital issue for AVs. The collision-free path planning problem in conditions with large road curvatures is investigated in this paper, with the consideration of environmental safety constraints, drivers’ comfort, vehicle actuator constraints, etc. Firstly, a Driver-Vehicle-Road (DVR) system is established based on the combination of the kinematic vehicle model and the two-point visual preview driver model, such that the driver's individual handling characteristics can be considered in the controller. The kinematic vehicle model is modified to have the similar understeering characteristics with those of the nonlinear full car models, and then the proposed DVR system can satisfy different groups of drivers and cars. Secondly, for environmental constraints, a new artificial potential field (APF) method is proposed, which can form a banana-shaped 3-D dangerous imaginary mountain and a lane boundary cliff suitable for arbitrary curvature roads to generate a collision-free evasive path. Finally, the Linear-Time-Varying (LTV) model predictive control (MPC) method is adopted to design the path planner. The CarSim-Simulink joint simulation illustrates that with the proposed planner, the host vehicle is capable of avoiding obstacles with a safer and more comfortable maneuver on large curvature roads. And the proposed path planner can provide individually safe trajectories for different drivers with good maneuverability. [ABSTRACT FROM AUTHOR]

Knowledge transfer is a promising concept to achieve real-time decision-making for autonomous vehicles. This paper constructs a transfer deep reinforcement learning (RL) framework to transform the driving tasks in the intersection environments. The driving missions at the unsignalized intersection are cast into a left turn, right turn, and running straight for automated vehicles. The goal of the autonomous ego vehicle (AEV) is to drive through the intersection situation efficiently and safely. This objective promotes the studied vehicle to increase its speed and avoid crashing other vehicles. The decision-making policy learned from one driving task is transferred through three transfer rules in another driving mission and evaluated. Simulation results reveal that the decision-making strategies related to similar tasks are transferable and have a high success rate. It indicates that the presented control framework could reduce time consumption and realize online implementation. Therefore, the transfer RL concept is helpful for establishing the real-time decision-making policy for autonomous vehicles. [ABSTRACT FROM AUTHOR]

To ensure trajectory tracking accuracy and driving safety, this paper presents a real time nonlinear model predictive control (NMPC) method for autonomous guided vehicles (AGVs) to avoid dangerous load transfer effect in carrying heavy equipment. The AGV with a four-wheel-steering four-wheel-driving (4WS4WD) structure is controlled with a steer and velocity input for each wheel. The tracking problem is implemented by the standard NMPC framework and the referenced trajectory is planned by the Reeds-Shepp algorithm. A Large load creates a considerable load transfer effect which is dangerous for the wheels; hence, the controller limits the lateral acceleration to reduce load transfer. To realize online numerical computation for NMPC, the continuation/generalized minimal residual (C/GMRES) method is applied. A semismooth transformation based on the Chen-Harker-Kanzow-Smale (CHKS) function is introduced to handle inequality constraints since C/GMRES only solves standard equality forms. Trajectory tracking simulations are carried out in contrast to the quadratic programming method without lateral load transfer optimization, and practical experiments on an AGV carrying 13 tons of steel are also performed. The results indicate the accuracy, efficiency, safety and applicability of the proposed method on heavy-duty AGVs. [ABSTRACT FROM AUTHOR]

The seven papers in this special section focuses on fog computing for autonomous and connected automobiles. With advances in wireless communications, machine learning, and sensing technologies, autonomous and connected cars are becoming a reality. Many potential applications (e.g., augmented reality for information providing through a heads-up display and accident avoidance in autopilot) require significant computing power to process data generated by vehicle sensors for near-real-time responses. Upgrading on-board computers is one option at relatively high cost. Another solution is cloud computing, where the traditional centralized approach suffers from long latency and unstable connections in a vehicular environment and may congest network backhaul with a large amount of data. [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]

Path tracking is one of the most crucial issues of autonomous vehicles, but uncertain factors including road roughness and wind gusts could cause lateral deviation and even divergence of the control system. Therefore, this paper proposes a hierarchical anti-disturbance tracking architecture based on the steer-by-wire (SBW) system to improve tracking accuracy and dynamic stability for autonomous vehicles. The proposed architecture is robust strongly against different types of disturbances in the path tracking process by means of hierarchical decoupling. Firstly, in the upper-layer path tracking module, a robust model predictive control (MPC) method with a feedback correction term calculates the expected front-wheel steering angle. Secondly, in the lower-layer SBW module, a nonlinear high gain rack force observer is proposed to compensate for the generalized rack force disturbance. Then, an improved non-singular fast terminal sliding mode auxiliary control law based on the super-twisting algorithm is presented to attenuate the norm-bounded rack disturbance effectively. Simulation and experiment results show that the proposed control architecture can improve the path tracking accuracy and has strong interference resistance and robustness against both modeled and unmodeled disturbances. [ABSTRACT FROM AUTHOR]

In this paper, we investigate a novel collision probability-aware human-machine cooperative planning and tracking method for enhancing safety of automated vehicles. Firstly, a long-term trajectory prediction is obtained by using Gaussian mixture models with vehicle historical data. After that, a novel risk assessment system based on the dynamic potential field (DPF) and the fuzzy inference system (FIS) is proposed to evaluate the risk level of the human driver's behaviors. Based on the assessed human driving risk, the human-machine cooperation is activated adaptively during the trajectory planning. A novel human-machine cooperative trajectory planning algorithm, named as HM- $p$ RRT, is proposed and used to incorporate the driver's intent and automation's corrective actions during trajectory planning. Testing results show that the proposed HM- $p$ RRT algorithm is able to simultaneously mitigate collision and provide a safe trajectory, effectively ensuring the safety of the vehicle and reducing conflicts during human-machine interactions. [ABSTRACT FROM AUTHOR]

Autonomous Vehicles ($AV$ s) take advantage of Machine Learning (ML) for yielding improved experiences of self-driving. However, large-scale collection of $AV$ s’ data for training will inevitably result in a privacy leakage problem. Federated Learning (FL) is proposed to solve privacy leakage problems, but it is exposed to security threats such as model inversion, membership inference. Therefore, the vulnerability of the FL should be brought to the forefront when applying to $AV$ s. We propose a novel Byzantine-Fault-Tolerant (BFT) decentralized FL method with privacy-preservation for $AV$ s called BDFL. In this paper, a Peer-to-Peer (P2P) FL with BFT is built by extending the HydRand protocol. In order to protect their model, each $AV$ uses the Publicly Verifiable Secret Sharing(PVSS) scheme, which allows anyone to verify the correctness of encrypted shares. The evaluation results on the MNIST dataset have shown that introducing decentralized FL into $ AV$ area is feasible, and the proposed BDFL is superior to other BFT-based FL method. Furthermore, the experimental results on KITTI dataset indicate the practicality of BDFL on improving performances of multi-object recognition in $ AV$ areas. Finally, the proposed PVSS-based data privacy preservation scheme is also justified its characteristic of no side-effect on models’ parameters by the experiments on the MNIST and KITTI datasets. [ABSTRACT FROM AUTHOR]

In the advanced driver assistance system (ADAS) and autonomous driving systems, accurate information on preceding vehicle motion states is vital for the path planning and control. To develop those intelligent driving systems, a modular integrated estimation algorithm for the preceding vehicle longitudinal and lateral states is proposed in this paper, the coupled nonlinear characteristics of vehicle dynamics are applied to improve state estimation accuracy. First, considering driver aggressiveness, a linear moving horizon estimator for the vehicle longitudinal speed is designed based on the car following model. Then, the estimated vehicle longitudinal speed is delivered to the lateral estimator module in real-time. For the arbitrary driving routes, the Serret-Frenet equations and nonlinear lateral dynamics are combined to describe the lateral motion of preceding vehicle. Furthermore, a nonlinear moving horizon estimator is constructed to estimate vehicle lateral states accurately. To reduce the computational burden, the multiple shooting (MS) method is introduced, then the objective function is divided into several segments to improve the strong nonlinearity produced by multiple iterations. The proposed estimation algorithm can effectively handle the initial biases, the additive biases and model error, which are common in the acquired signals, and provide an accurate estimate of preceding vehicle states. [ABSTRACT FROM AUTHOR]

In the last years, more and more research activities have focused on the application of Virtual Reality (VR) and related cutting-edge immersive technologies like Augmented Reality (AR) and Mixed Reality (MR) to vehicular systems for supporting the design of novel Human-Machine Interaction techniques, simulating the functioning of newly developed components, recreating hazardous test environments, establishing trust in Connected and Autonomous Vehicles (CAVs), etc. The benefits brought by immersive environments for supporting simulations in the field of vehicular technology have already been recognized by many companies and academic institutions worldwide, as confirmed by the huge investments and a large number of publications. As we approach the time in which immersive simulations will become fundamental for a wide range of applications in the field of vehicular technology, the goal of this Special Section was to attract high-quality submissions reporting state-of-the-art research activities targeted to next-generation vehicles and related infrastructures. The Special Section received 12 submissions, which were prescreened and reviewed by experts in the field. Four papers were ultimately accepted. [ABSTRACT FROM AUTHOR]

Scene flow represents the 3D motion of every point in the dynamic environments. Like the optical flow that represents the motion of pixels in 2D images, 3D motion representation of scene flow benefits many applications, such as autonomous driving and service robot. This paper studies the problem of scene flow estimation from two consecutive 3D point clouds. In this paper, a novel hierarchical neural network with double attention is proposed for learning the correlation of point features in adjacent frames and refining scene flow from coarse to fine layer by layer. The proposed network has a new more-for-less hierarchical architecture. The more-for-less means that the number of input points is greater than the number of output points for scene flow estimation, which brings more input information and balances the precision and resource consumption. In this hierarchical architecture, scene flow of different levels is generated and supervised respectively. A novel attentive embedding module is introduced to aggregate the features of adjacent points using a double attention method in a patch-to-patch manner. The proper layers for flow embedding and flow supervision are carefully considered in our network designment. Experiments show that the proposed network outperforms the state-of-the-art performance of 3D scene flow estimation on the FlyingThings3D and KITTI Scene Flow 2015 datasets. We also apply the proposed network to the realistic LiDAR odometry task, which is a key problem in autonomous driving. The experiment results demonstrate that our proposed network can outperform the ICP-based method and shows good practical application ability. The source codes will be released on https://github.com/IRMVLab/HALFlow. [ABSTRACT FROM AUTHOR]

Priority and mutual interaction among vehicles are crucial for the efficiency and safety control of unsignalized intersections. However, most of the approaches for intersection cooperation are based on first-come-first-served (FCFS) rules, neglecting the influence caused by different priorities. Focusing on the priority and mutual interaction between connected and automated vehicles (CAVs), this paper proposes a game-based hierarchical control strategy to improve the traveling efficiency at unsignalized intersections. The three-layered hierarchical strategy includes the priority negotiation layer (PNL), strategy bargaining layer (SBL) and strategy optimization layer (SOL), respectively. In the PNL, a game-theoretic strategy is designed for priority negotiation, in which the most efficient crossing order is sought by the Pareto-optimal set. In the SBL, with the assigned negotiated priority, a bargaining game is designed for further interaction between CAVs. In the SOL, model predictive control is applied for strategy optimization considering the comfort of drivers and executability of vehicles. Two cases of cooperative intersection, in which fixed priority with comparison of the Nash game method and greedy method, and negotiated priority with comparison of fixed priority are carried out in a hardware-in-the-loop (HiL) test. The results demonstrate that the traffic efficiency is improved with safety guaranteed by the proposed hierarchical strategy. [ABSTRACT FROM AUTHOR]

With the increase in the capacity and power rating of batteries in today's HEVs and EVs, the battery thermal management (BTM) system bears growing importance in vehicle safety and efficiency. A practical BTM system is commonly coupled with the passenger cabin heating-ventilation-air-conditioning (HVAC) system, which makes a major energy consumer and a challenging control object. Thanks to the connected and automated vehicle (CAV) technology, predictions of the vehicle speed profile and power trajectory can be obtained, providing the possibility for predictive control of the BTM and HVAC coupled system in order to maintain battery safety, passenger comfort, and to reduce energy consumption. However, the tradeoff among higher energy saving potential and wider control range from a long and sparse horizon, and higher accuracy from a short and dense horizon is inevitable in the conventional predictive control. In this paper, a two-layer predictive control strategy for warm/hot weather is proposed to address the aforementioned tradeoff. The upper layer controller firstly plans the optimized battery temperature trajectory according to the look-ahead speed preview and battery power profiles provided by the CAV network and the integrated BTM and HVAC system efficiency surface from off-line data. Then the lower layer model predictive controller tracks the planned trajectory and cabin temperature reference while enforcing the energy consumption optimization. Simulation results demonstrate that the proposed strategy exhibits accurate battery and passenger cabin temperature reference tracking while achieving up to 10.47% HVAC energy saving comparing to a baseline control strategy with UDDS profile and typical driving conditions. Implementation of the proposed strategy on a real-time vehicle emulator based on rapid control prototyping (RCP) and hardware-in-the-loop (HIL) platforms demonstrates the real-world implementation capability of the proposed two-layer framework. [ABSTRACT FROM AUTHOR]

Recently, inferring lane change intention has received considerable attention. Due to the high nonlinearity and complexity of traffic contexts, traditional methods cannot satisfy the requirements of long-term prediction tasks and lack the ability of capturing nonlinear temporal dependencies. This paper proposes an intention inference model based on Recurrent Neural Networks (RNN), to tackle time series prediction problems. Considering dynamic interaction among surrounding vehicles, our model takes the sequence motion information of surrounding vehicles as inputs and calculates the congestion of different lanes, integrated with vehicle states of the object vehicle. To illustrate the availability of the proposed RNN intention inference model, a motion planning controller considering intention was developed. A Nonlinear Model Predictive Control (NMPC) was established to planning a safe, sub-optimal path for autonomous driving vehicle under collision avoidance constraints. The experiments on the proposed model were conducted, based on two RNN structure Long-Short Term Memory (LSTM) and Generalized Recurrent Unit (GRU), by Tensorflow with NGSIM data. The motion planning controller is modeled and simulated by Carsim with Simulink for some typical scenarios. Subsequently, experimental results demonstrate that RNN achieves best performance, inferring intention with 96% accuracy, compared with other approaches. [ABSTRACT FROM AUTHOR]

This paper addresses vehicle positioning, a topic whose importance has risen dramatically in the context of future autonomous driving systems. While classical methods that use GPS and/or beacon signals from network infrastructure for triangulation tend to be sensitive to multi-paths and signal obstruction, our method exhibits robustness with respect to such phenomena. Our approach builds on the recently proposed Channel-SLAM method which first enabled leveraging of multi-path so as to improve (single) vehicle positioning. Here, we propose a cooperative mapping approach which builds upon the Channel-SLAM concept, referred to here as Team Channel-SLAM. Team Channel-SLAM not only exploits the stationary nature of many reflecting objects around the vehicle, but also capitalizes on the multi-vehicle nature of road traffic. The key intuition behind our method is the exploitation for the first time of the correlation between reflectors around multiple neighboring vehicles. An algorithm is derived for reflector selection and estimation, combined with a team particle filter (TPF) so as to achieve high precision simultaneous multiple vehicle positioning. We obtain large improvement over the single-vehicle positioning scenario, with gains being already noticeable for moderate vehicle densities, such as over 40% improvement for a vehicle density as low as 4 vehicles in 132 meters’ length road. [ABSTRACT FROM AUTHOR]

Transfer learning enables to re-use knowledge learned on a source task to help learning a target task. A simple form of transfer learning is common in current state-of-the-art computer vision models, i.e., pre-training a model for image classification on the ILSVRC dataset, and then fine-tune on any target task. However, previous systematic studies of transfer learning have been limited and the circumstances in which it is expected to work are not fully understood. In this paper we carry out an extensive experimental exploration of transfer learning across vastly different image domains (consumer photos, autonomous driving, aerial imagery, underwater, indoor scenes, synthetic, close-ups) and task types (semantic segmentation, object detection, depth estimation, keypoint detection). Importantly, these are all complex, structured output tasks types relevant to modern computer vision applications. In total we carry out over 2000 transfer learning experiments, including many where the source and target come from different image domains, task types, or both. We systematically analyze these experiments to understand the impact of image domain, task type, and dataset size on transfer learning performance. Our study leads to several insights and concrete recommendations: (1) for most tasks there exists a source which significantly outperforms ILSVRC’12 pre-training; (2) the image domain is the most important factor for achieving positive transfer; (3) the source dataset should include the image domain of the target dataset to achieve best results; (4) at the same time, we observe only small negative effects when the image domain of the source task is much broader than that of the target; (5) transfer across task types can be beneficial, but its success is heavily dependent on both the source and target task types. [ABSTRACT FROM AUTHOR]

Good responsiveness of autonomous vehicle computing systems is crucial to safety and performance of the vehicles. For example, an autonomous vehicle (AV) may cause an accident if the end-to-end latency from sensing a pedestrian to emergency stop is too high. However, the AV software stacks are too complex to probabilistically analye the end-to-end latency on a multi-core system. They consist of a graph of tasks with different periods, and have a large variability in the task execution times, which may lead to the maximum core utilization $U^{\max }$ U max greater than 1.0 on some cores. This paper proposes a novel stochastic analysis of the end-to-end latency over the AV stacks that allows $U^{\max }$ U max to exceed 1.0 on each core. The proposed analysis models the entire stack as a graph of task graphs under a multi-core partitioned scheduling and provides a probabilistic guarantee that the analyzed latency distribution upper-bounds the one observed from a real system under the assumption of independent task execution times. Using the Autoware stack with inter-task dependent execution times, it is shown that our analysis, combined with a task grouping to mitigate the inter-task correlations, can give a latency distribution for each task path that almost upper-bounds the observed one. [ABSTRACT FROM AUTHOR]

This paper focuses on the controller design for path-tracking problem of autonomous ground vehicles (AGVs) by employing a multi-constraint nonlinear predictive control (NMPC) schema. It is aimed to improve the transient performance of the vehicle and to consider a rollover prevention criterion in the proposed method. The path-tracking problem is transformed into the yaw stabilization issue, and a feedback control law with input saturation is developed to decrease the steady-state errors. Furthermore, the yaw-rate signal is generated for the desired path-tracking performance. The major contributions of the present paper are, first, developing a neural network autoregressive with exogenous input system to assist in obtaining an accurate and explicit model in order to contribute to the control of the system over the prediction horizon; second, describing a Frenet–Serret differential geometry based path-following agenda and developing AGV dynamic model by incorporating the vehicle vertical mode of motion to prevent vehicle rollover during critical maneuvers, and finally, achieving an enhanced yaw stabilization and transient tracking performance considering saturated input signal by employing the proposed system identification algorithm. The effectiveness of the proposed control system is verified by comparing with the traditional NMPC method by employing a high fidelity CarSim/MATLAB framework. [ABSTRACT FROM AUTHOR]

This paper investigates the path-tracking control issue for autonomous ground vehicles with the integral sliding mode control (ISMC) considering the transient performance improvement. The path-tracking control is converted into the yaw stabilization problem, where the sideslip-angle compensation is adopted to reduce the steady-state errors, and then the yaw-rate reference is generated for the path-tracking purpose. The lateral velocity and roll angle are estimated with the measurement of the yaw rate and roll rate. Three contributions have been made in this paper: first, to enhance the estimation accuracy for the vehicle states in the presence of the parametric uncertainties caused by the lateral and roll dynamics, a robust extended Kalman filter is proposed based on the minimum model error algorithm; second, an improved adaptive radial basis function neural network (RBFNN) considering the approximation error adaptation is developed to compensate for the uncertainties caused by the vertical motion; third, the RBFNN and composite nonlinear feedback (CNF) based ISMC is developed to achieve the yaw stabilization and enhance the transient tracking performance considering the input saturation of the front steering angle. The overall stability is proved with Lyapunov function. Finally, the superiority of the developed control strategy is verified by comparing with the traditional CNF with high-fidelity CarSim-MATLAB simulations. [ABSTRACT FROM AUTHOR]

This paper studies the codesign optimization approach to determine how to optimally adapt automatic control of an intelligent electric vehicle to driving styles. A cyber-physical system (CPS)-based framework is proposed for codesign optimization of the plant and controller parameters for an automated electric vehicle, in view of vehicle's dynamic performance, drivability, and energy along with different driving styles. System description, requirements, constraints, optimization objectives, and methodology are investigated. Driving style recognition algorithm is developed using unsupervised machine learning and validated via vehicle experiments. Adaptive control algorithms are designed for three driving styles with different protocol selections. Performance exploration method is presented. Parameter optimizations are implemented based on the defined objective functions. Test results show that an automated vehicle with optimized plant and controller can perform its tasks well under aggressive, moderate, and conservative driving styles, further improving the overall performance. The results validate the feasibility and effectiveness of the proposed CPS-based codesign optimization approach. [ABSTRACT FROM AUTHOR]