High-speed autonomous truck platooning with short inter-vehicle spacing is essential for improving freight efficiency. However, in mixed traffic environments, safety and stability are strongly affected by sudden human-driven disturbances and stochastic communication delays. This study proposes a two-layer Koopman-based framework that combines open-loop and closed-loop data-driven predictors with delay-aware compensation and safety assurance for distributed platoon control. In the first layer, an open-loop Koopman predictor is integrated with a robust control barrier function based safety filter formulated as a constrained quadratic optimization. This formulation ensures that nonlinear vehicle dynamics and model uncertainty are handled with explicit safety margins. In the second layer, a closed-loop Koopman predictor performs forward state estimation to compensate delayed predecessor information and reduce delay-induced observation bias in the control loop. The method is evaluated in three representative scenario classes, with random communication delays up to 500 ms and platoon sizes from 3 to 10 vehicles. Under uniformly distributed 300 ms delay, the proposed strategy limits the maximum absolute spacing error to 0.8 m (versus 5.5 m for baseline MPC and 1.94 m for robustH∞), while preserving string-stable car-following behavior. In emergency sudden-braking scenarios, it increases the minimum following distance to 5.9 m, compared with 1.3 m without a safety filter and 2.4 m with a linear control barrier function, demonstrating effective safety correction under critical disturbances.

Deep Reinforcement Learning (DRL) is becoming a prominent method for autonomous driving due to its strong capability to generate complex driving policy. However, DRL motion planning still has limitations in safety performance including learning quality, convergence speed and the safety guarantee. To this end, this work proposes a safety-enhanced deep reinforcement learning method with dynamic safety guidance (DSG-DRL) for lane-change motion planning. It bears the following key features: 1) Able to learn a safer DRL driving policy by additionally including potentially unsafe behaviors; 2) Able to accelerate learning a safe policy by making dangerous driving experiences impressive; 3) Able to further enhance the driving safety by avoiding unexpected reckless action. The proposed DSG-DRL motion planner dares to make mistakes to learn the safe driving policy better and faster. By evaluating anticipated risk, it learns not only from the maneuvers right at the moments of collisions, but also from the dangerous maneuvers leading towards collisions. Besides, risk driving experiences are enhanced with additional memory batches and sampling prioritization. Moreover, reckless actions can be prevented by dynamic constraints both in training and testing, which further improves the safety performance. Simulation validation shows that the proposed method can learn a safer driving policy with faster convergence speed, achieving the high safety performance while keeping the driving efficiency.

This study addresses the resilient control of connected and automated vehicle (CAV) platoons under denial of service (DoS) attacks. We investigate two main attack scenarios. In the first scenario, zero-input DoS attacks target the communication channels, and progressing from degraded vehicle communication ability to compromised leader reachability may cause complete platoon disconnection. The second scenario considers the simultaneous occurrence of the above zero- and hold-input DoS attacks, where the latter introduces control delay. With Lyapunov functions and dwell time methods, switching distributed controllers are developed to achieve input-to-state string stability (ISSS). This ensures robust platoon performance through spacing regulation, speed consensus, and state error attenuation along the platoon, even under different attack conditions. Numerical simulations demonstrate the platoon’s resilience against simultaneous zero-input and hold-input DoS attacks, highlighting their cascading effects on platoon performance.

Autonomous navigation technology for autonomous ground vehicles (AGVs) is currently a highly active research area. With the advancement of internet of things (IoT) technologies, AGVs increasingly leverage interconnected systems, such as onboard sensors, vehicle-to-everything (V2X) communication, and cloud data sharing, to enhance navigation capabilities. Human-in-the-loop (HIL) guidance has been shown to be effective in improving the performance of reinforcement learning (RL) algorithms. Existing methods in this field often assume that human guidance is always beneficial. However, incorrect guidance can cause oscillations or even divergences in RL training. In this study, we propose an innovative trust-calibrated HIL-RL approach to address these gaps. First, human guidance is introduced into the RL framework to enhance learning performance through intervention and demonstration. This process includes adding behavior cloning (BC) objectives to the RL policy and an adaptive experience replay mechanism. Second, a trust evaluation mechanism is incorporated within the HIL-RL framework to calculate a belief value, which not only ensures that human guidance is trustworthy but also dynamically optimizes the BC weight. This improvement enhances the training efficiency of RL agents and supports steady performance improvement, even when exposed to potentially detrimental external intervention. The results in simulation show that the proposed method achieves an improvement in success rate of 22–26% over vanilla RL. In real-world experiments, the proposed method achieved a 100% success rate and demonstrated outstanding navigation efficiency, validating the effectiveness of the trust evaluation mechanism.

Intelligent Automation Systems, such as automated storage and retrieval systems, self-driving vehicles, various types of autonomous robots, and autonomous workshop plants, act independently of direct human supervision. Their impact on society and human life will be significant. These automation systems are safety-critical, complex, and powerful with a higher-level functionality. Safety and reliability to perform their tasks safely and minimize failures is one of the key challenges and becomes costly and difficult to achieve. The development of novel safety and reliability technologies dealing with theoretical aspects and for intelligent automation systems has become a hot spot in recent years. With the novel safety and reliability technologies, the intelligent automation systems can detect system failures, identify operation risks, predict unknown safety hazards and vulnerabilities, and avoid situations that pose risks to humans, property, or the automation systems themselves. Recent developments confirm that there are still areas of research to be explored within the safety-critical approaches. This Special Issue on Emerging Trends in Safety-critical Issues for Intelligent Automation Systems of IEEE Transactions on Automation Science and Engineering (TASE) aims to present recent advances in theories, methods, and applications that address safety-critical challenges in autonomous intelligent systems. The objective is to compile state-of-the-art research that contributes to the development of resilient and trustworthy automation solutions in safety-sensitive scenarios. After a thorough and rigorous peer-review process, 27 high-quality articles were selected from numerous submissions worldwide. These articles closely align with the Special Issue’s scope and can be categorized into seven key topics. - Risk assessment and model-based safety and cybersecurity analysis [A1], [A2], [A3], [A4], [A5]. - Intelligent fault detection and fault-tolerant control [A6], [A7], [A8], [A9], [A10]. - Reliability and traceability of decision-making for intelligent automation systems [A11], [A12], [A13]. - Conflict detection and resolution in intelligent automation systems [A14], [A15]. - Safety- and security-related issues including functional safety and system security [A16], [A17], [A18]. - Design, development, validation, and applications of intelligent automation systems such as UAVs, UGVs, and UUVs [A19], [A20], [A21], [A22], [A23]. - Human-robot collaboration, risk assessment of intelligent automation [A24], [A25], [A26], [A27]. We believe this collection will provide a valuable reference for academia and industry alike, and inspire further research in this rapidly evolving field.

The paper studies an Internet of Vehicles system consisting of several UAVs sending downlink information to ground vehicles to move an uneven terrain. In such environments, the transmission from the UAV to the ground vehicles may be blocked by tall buildings or mountains. Our goal is to construct multi-UAV navigation laws, transmission schedules, and transmit power levels of the UAVs that maximize the total downlink throughput of the system while minimizing the total transmit power and the propulsion energy of the multi-UAV team subject to a communication interference constraint for the UAVs. An important feature of the investigation is that a safety constraint that guarantees to avoid collisions between UAVs with a required safety margin should always be satisfied. A real-time law for joint communication and multi-UAV navigation is proposed. A mathematically rigorous analysis of the developed algorithm and proof of its optimality are given. Illustrative examples and simulations demonstrate the effectiveness of the approach. The main purpose of the proposed framework is to achieve joint communication and safe 3D path optimization, which means that the objective is to jointly construct 3D paths of the UAVs, transmission schedules, and transmit power levels of the UAVs. An advantage of this approach is that it allows to simultaneously maximize the total downlink throughput of the system and minimizes the total transmit power and the propulsion energy of all the UAVs subject to safety constraints that guarantee the UAVs avoid collisions between the UAVs as well as the UAVs and the uneven terrain.Note to Practitioners—Unmanned aerial vehicles (UAVs) have been regarded as a promising platform to serve as aerial access points, especially in complex environments with tall buildings and uneven terrains where conventional ground access points might usually provide unqualified service to users. In this paper, we have investigated an Internet of Vehicles (IoV) system that involves multiple unmanned aerial vehicles (UAVs) transmitting downlink information to ground vehicles navigating through challenging terrains. Our primary objective is to develop effective multi-UAV navigation laws, transmission schedules, and transmission power allocations that optimize the total downlink throughput of the system. At the same time, we aim to minimize the overall transmit power and propulsion energy of the UAV team, while adhering to communication interference constraints. It is worth noting that ensuring safety is of utmost importance in our investigation. We have incorporated a safety constraint that guarantees the prevention of collisions between UAVs, maintaining a required safety margin at all times. To tackle these challenges, we propose a real-time law that combines dynamic programming and model predictive control techniques for joint transmission and multi-UAV trajectory planning. The developed algorithm has undergone a rigorous mathematical analysis, and we have provided proof of its optimality.

Due to the complexity and variability of real-world environments, data-driven autonomous navigation strategies for autonomous ground vehicles have significant potential to improve performance and adaptability in diverse scenarios. Reinforcement learning (RL) has emerged as a promising approach for autonomous navigation. However, existing RL methods often struggle with low sample efficiency, limited adaptability, and poor generalization in dynamic multi-task scenarios. To address these issues, we propose a novel framework: human-in-the-loop adversarial RL with diffusion policy, designed for scalable and robust policy learning. This framework leverages a diffusion model as policy network, effectively exploring and learning high-dimensional, multi-modal behavior distributions. It also integrates human feedback to improve data efficiency and stabilize policy training. On top of this, adversarial training is employed to improve robustness and adaptability to change in tasks and distributions. The proposed method is trained in simulation, and then the well-trained policy is transferred to the real-world. Experimental results demonstrate that this approach significantly outperforms existing methods in terms of efficiency, stability, generalization, and multi-task adaptability, offering a promising solution for the next generation of autonomous navigation systems. The supplementary video is available at https://youtu.be/JH3knw0I5lU

This study proposes a novel distributed model predictive control (DMPC) approach for mixed vehicle platoons (MVPs), which achieves driving safety, asymptotic stability, and traffic wave dissipation simultaneously. The longitudinal dynamics model and safety constraints are first established for each vehicle. The MVPs are divided into several sub-platoons according to the distribution of connected-and-automated vehicles (CAVs) and human-driven vehicles (HDVs). Then, a compound controller to ensure the head-to-tail string stability of MVPs is constructed as a reference controller for the subsequent design of the double-layer distributed model predictive controller (DL-DMPC). To describe the behavior of human drivers, a car-following model specific to HDVs is developed. On the basis of the designed compound controller and the description of HDVs, the DL-DMPC is proposed. The first layer improves tracking performance and satisfies the state constraints of each CAV under different communication topologies through an optimization problem. The second layer utilizes controller-matching technology to ensure the asymptotic stability of vehicle platoons and dissipate traffic waves. With the above double-layer structure, the DL-DMPC can simultaneously address multiple tasks and is applicable in various communication topologies. Simulations and analyses based on the NGSIM dataset are conducted in various scenarios to validate the effectiveness of the developed approach.

The development of autonomous vehicles has shown great potential to enhance the efficiency and safety of transportation systems. However, the decision-making issue in complex human-machine mixed traffic scenarios, such as unsignalized intersections, remains a challenge for autonomous vehicles. While reinforcement learning (RL) has been used to solve complex decision-making problems, existing RL methods still have limitations in dealing with cooperative decision-making of multiple connected autonomous vehicles (CAVs), ensuring safety during exploration, and simulating realistic human driver behaviors. In this paper, a novel and efficient algorithm, Multi-Agent Game-prior Attention Deep Deterministic Policy Gradient (MA-GA-DDPG), is proposed to address these limitations. Our proposed algorithm formulates the decision-making problem of CAVs at unsignalized intersections as a decentralized multi-agent reinforcement learning problem and incorporates an attention mechanism to capture interaction dependencies between ego CAV and other agents. The attention weights between the ego vehicle and other agents are then used to screen interaction objects and obtain prior hierarchical game relations, based on which a safety inspector module is designed to improve the traffic safety. Furthermore, both simulation and hardware-in-the-loop experiments were conducted, demonstrating that our method outperforms other baseline approaches in terms of driving safety, efficiency, and comfort.

Accurate and real-time acquisition of vehicular system dynamic states, road surface conditions, and motion states of surrounding participants is crucial for the safety, passenger comfort, and operational efficiency of autonomous vehicles (AVs) and connected automated vehicles (CAVs). In recent years, a significant amount of research has contributed to the field of state estimation for vehicles, roads, and pedestrians. From the systemwide perspective of intelligent transportation systems to a focused view on “vehicle-road-pedestrian”, this survey aims to provide a comprehensive review and summary of recent state estimation techniques for vehicle motion, road surface, and pedestrian motion. A thorough analysis of the reviewed literature, relevant datasets, evaluation metrics, and experimental platforms in this field is also conducted. Finally, existing challenges and future research directions about methods and performance evaluation are further discussed. This survey is expected to contribute to the advancement of research in dynamic state estimation of vehicle-road-pedestrian, thereby facilitating the development of efficient and safe intelligent transportation systems.

This work focuses on the yaw stabilization of the four-wheel-independent-drive electric vehicle (FWID-EV) with the constrained active front steering (AFS) and direct yaw-moment control (DYC). First, a modified tire model is employed in the design of the unscented Kalman filter to realize the estimation of the tire-road friction coefficient (TRFC), and a backpropagation neural network is developed to online estimate the tire cornering stiffness; Second, a yaw stabilization supervisor is designed to solve the conflicts between the AFS and DYC systems, and the mode-boundary maps of the tire operating regions are utilized to generate the triggered signals so as to activate the systems; Third, a hierarchical adaptive model predictive control (MPC), including the estimation, activation, compensation, and distribution layers is proposed for yaw stabilization of the FWID-EV under the extreme conditions. Emergency maneuvers under big path curvature, low TRFC, and high vehicle speed are designed. Both software-in-the-loop and hardware-in-the-loop tests are performed to examine the effectiveness and practicability of the proposed methods, respectively.

This study explores the effect of takeover time (TOT) on decision-making for Level-3 autonomous vehicles (L3-AVs). The existing research on L3-AV lacks an in-depth analysis of the mechanisms affecting TOT, ignores the importance of spatial and temporal variations in features for TOT prediction, and also lacks consideration of TOT in downstream trajectory planning tasks. This study proposed an exponential smoothing transformers (ETS) former model for TOT prediction, and then, the spatial-temporal predictive transformer (ST-Preformer) was employed to forecast the trajectories of surrounding vehicles, assess lane availability, and determine lane-changing probabilities. Ultimately, these evaluations contribute to the decision-making process of L3-AVs. The findings showed that the ETSformer was able to explain more than 83% of the characteristics of the TOT distribution in the TOT prediction task, effectively reducing the absolute percentage error by 0.7%, based on which the decision-making framework was able to make safe and comfortable optimal decisions. Decision-making is closely related to driving conditions and the surrounding traffic state, and TOT has a critical impact on the safety and stability of decision-making. A comprehensive understanding the impact of TOT on decision-making can help improve the safety of autonomous driving and provide guidance for improving decision-making techniques.

Given the limitations of current methods in terms of accuracy and efficiency for robot scene recognition (SR) in domestic environments, this paper proposes an active scene recognition approach (ASR) that allows the robot to recognize scenes correctly using less images, even when the robot's position and observation direction are uncertain. ASR includes a behavior cloning-based action classification model, which can adjust the robot view actively to capture beneficial images for scene recognition. To address the lack of essential expert data for training the action model, we introduce an expert data generation method that avoids time-consuming and inefficient manual data collection. Additionally, we present a multi-view scene recognition method to handle the multiple images resulting from view changes. This method includes a scene recognition model that scores each image and a revision and prediction method to mitigate the compounding error introduced by behavior cloning as well as output the finial recognition result. We conducted numerous comparative experiments and an ablation study in various domestic environments using a publicly simulated platform to validate our ASR method. The experimental results demonstrate that our proposed approach outperforms state-of-the-art methods in terms of both accuracy and efficiency for scene recognition. Furthermore, our method, trained in simulated environments, demonstrates excellent generalization capabilities, allowing it to be directly transferred to the real world without the need for fine-tuning. When deployed on a TurtleBot 4 robot, it achieves precise and efficient scene recognition in diverse real-world environments.

To enhance path tracking performance in the presence of steering motor faults, this paper introduces an active fault-tolerant control strategy for electric vehicles with four inwheel motors. Firstly, the single-track vehicle dynamics model and steering faults model are established. The control framework includes an upper-level linear parameter-varying model predictive controller for active front steering, an upper-level event-triggered predictive controller for direct yaw control, and a lower-level torque allocation controller. The bi-directional long short-term memory (Bi-LSTM) network is used for low-latency rapid detection of steering system faults. If the fault is detected, the upper-level controller for direct yaw control is triggered to mitigate the negative impact of the steering actuator faults. Based on the high-fidelity CarSim model, the simulation test is conducted under a double-lane change scenario with severe stuck faults in the steering system. The simulation results indicate that the proposed scheme can reduce the cumulative tracking error by 37.86% under the set stuck faults compared with the baseline method.

In this paper, we propose the innovative use of Intelligent Vehicle Carriers (IVCs) as a key solution to address the energy constraints of small-scale unmanned Intelligent Vehicles (IVs). IVCs function as both transporters and charging stations, significantly boosting the operational range and efficiency of IVs. Our research delves into the IV-IVC collaborative framework, highlighting the existing challenges, exploring potential solutions, and examining a range of applications. This study offers a visionary approach to revolutionizing intelligent transportation systems by leveraging the synergistic relationship between IVs and IVCs.

Internet of Things (IoT) devices have been widely deployed to build smart cities. How to efficiently collect data from large-scale IoT devices is a valuable and challenging research topic. Benefiting from agility, flexibility, and deployability, an unmanned aerial vehicle (UAV) has great potential to be an aerial base station. However, given the limited battery capacity, the flight time of a UAV is limited. This article focuses on using multi-UAVs to execute long-distance data collection from large-scale IoT devices. We design a multi-UAVs-assisted large-scale IoT data collection system. The core facilities of this system are the data center and charging stations, which are equipped with a limited number of charging piles to provide charging services for UAVs. To ensure the efficient operation of the system, the problem of deployment and flight planning of UAVs is formulated as a joint optimization problem. To solve the problem, a population-based optimization algorithm with a three-layer structure, namely, EDDE-DPDE, is proposed. It includes two core components: 1) elite-driven differential evolution (EDDE) and 2) differential evolution with a dynamic population (DPDE), which are two variants of differential evolution. Thanks to ideas of reusing elite individuals and historical information, the proposed EDDE-DPDE shows an improvement of at least 11.11% compared with four powerful algorithms in terms of average travel time.

Despite recent advancements in robotic exploration, estimating the three-dimensional (3-D) motion of unknown dynamic objects, which are prevalent in chaotic construction sites, remains an unresolved challenge. In this work, we studied the problem of detecting and tracking the six-degree-of-freedom (6-DoF) motion of unknown moving objects using only a stereo camera. The fundamental idea of our approach is to estimate the positions of map points at each timestamp and model their uncertainty in position, such that the correlation between map points can be found by segmenting point clouds that are parts of moving objects into different groups. By analyzing the correlation between map points, we can detect the dynamic objects in the scene without making any assumptions about the type of objects. Thus, this approach enables tracking of both known and unknown moving objects, such as a robot carrier with stacked luggage. It surpasses the performance of existing appearance-based methods, which often face difficulties when dealing with unknown objects. Through extensive real-world experiments, we demonstrate the effectiveness of our approach in accurately tracking moving objects, highlighting its potential for various applications. In addition, we successfully deploy our object motion estimation algorithm in an unmanned ground vehicle (UGV) for the purpose of avoiding unknown moving objects in real-world scenarios. This practical implementation underscores the applicability of our approach in real-world settings.

This paper introduces a novel algorithm, refer to NEWDQN, which is based on the deep Q-network (DQN) framework. The primary objective of this algorithm is to optimize the successful rate both in autonomous drone obstacle avoidance and target tracking tasks, while this algorithm can also improve the drawbacks of the previous algorithm in convergence. Furthermore, the algorithm endows the drone with environment perception capabilities and incorporates a direction-based reward-penalty function into the reward function, enhancing the drone's generalization ability and overall performance. Extensive simulations demonstrate that compared to conventional DQN and Double DQN (DDQN) algorithms, NEWDQN exhibits faster convergence speed, shorter tracking paths, and more robust adaptability to different environments.

This article addresses a joint path planning and communication scheduling problem for a team of unmanned aerial vehicles (UAVs) equipped with central processing units and performing computing tasks for Internet of Things (IoT) devices located on uneven terrain. On such terrains, the Line of Sight (LoS) between a UAV and a ground IoT device may be blocked by buildings or mountains. In this article, reconfigurable intelligent surfaces (RISs) are deployed on the ground to improve wireless communication between UAVs and IoT devices. Computing tasks and their results can be sent to and from UAVs either directly or via RIS-reflected paths. The objective is to develop a joint multi-UAV path planning/transmission scheduling algorithm that maximizes the number of computing tasks successfully and timely performed by the UAVs and transmitted back to the IoT ground devices and minimizes the total energy consumption of the UAVs. An effective path planning algorithm for this optimization problem is proposed. A mathematically rigorous proof of its asymptotic optimality is given. The developed algorithm is computationally effective and easily implementable in real time. Computer simulations and comparisons with other effective methods prove the effectiveness of the developed approach.

This paper considers using drones to conduct the last-mile parcel delivery. To enable the beyond-battery-lifetime flight, drone stations are considered to replace or recharge the battery for drones. We focus on the flight planning problem with the goal of minimizing the total travel time from the depot to a customer, a key indicator of the quality of service. We investigate four typical ways for the drone to get extra energy at drone stations: 1) replacing the battery with a fresh one, 2) recharging the battery to the full capacity, 3) recharging the battery to the optimal level, and 4) recharging the battery to the optimal level accounting for the availability of drone stations (i.e., whether a drone station is occupied by other drones). While the first two scenarios can be formulated following the framework of integer linear programming, the last two scenarios turn into mixed-integer nonlinear programming problems. To address the later problems, we present a framework in which discretized state graphs are constructed first and then the optimal paths are found by graph searching algorithms. We propose a dynamic version of Dijkstra’s algorithm to deal with the unavailability issue of drone stations. The algorithm can quickly find the optimal flight path for a drone, and extensive computer-based experimental results have been presented to demonstrate the effectiveness of the proposed method. Note to Practitioners—Multi-rotary unmanned aerial vehicles (UAVs), also known as drones, have been regarded as a promising means to reshape future logistics. To save human labour and reduce cost, many giant logistics companies have been dedicated to developing various drones to deliver light and small parcels during the past decade. However, due to the limitation of payload, the battery capacity is constrained, which prevents drones from long-distance flights. Practitioners have tried the drone-vehicle collaboration method, but this still requires human labour to participate. In this paper, we present a framework where drones autonomously conduct long-distance delivery with the assistance of drone stations. It is worth pointing out that such a framework is not to replace the ground delivery method but to serve as an alternative to the ground counterpart for small and light parcels. A particular focus is on the flight planning from the depot to a destination, which includes not only a sequence of drone stations to stop at but also the corresponding rest time to recharge the battery. Several typical scenarios about battery recharging are discussed, and a dynamic version of Dijkstra’s algorithm is presented to deal with the challenging case where drone station resources are limited. The presented approach is able to find out the optimal flight plan quickly.

Using drones to carry out commercial parcel delivery can significantly promote the transformation and upgrading of the logistics industry thanks to the saving of human labor source, which is becoming a new component of intelligent transportation systems. However, the flight distance of drones is often constrained due to the limited battery capacity. To address this challenge, this paper designs a multi-drones-assisted commercial parcel delivery system, which supports long-distance delivery by a generalized service network (GSN). Each node of the GSN is equipped with charging piles to provide a charging service for drones. Given the limited number of charging piles at each node and the limited battery capacity of a drone, to ensure the efficient operation of the system, the flight planning problem of drones is converted into a large-scale optimization problem by a priority-based encoding mechanism. To solve this problem, an enhanced backtracking search algorithm (EBSA) is reported, which is inspired by the characteristics of the considered flight planning problem and the weak ability of the backtracking search algorithm to escape from a local optimum. The core components of EBSA are the designed comprehensive learning mechanism and local escape operator. Experimental results prove the validity of the improved strategies and the excellent performance of EBSA on the considered flight planning problem.

Interest in autonomous driving (AD) and intelligent vehicles (IVs) is growing at a rapid pace due to the convenience, safety, and economic benefits. Although a number of surveys have reviewed research achievements in this field, they are still limited in specific tasks and lack systematic summaries and research directions in the future. Our work is divided into three independent articles and the first part is a survey of surveys (SoS) for total technologies of AD and IVs that involves the history, summarizes the milestones, and provides the perspectives, ethics, and future research directions. This is the second part (Part I for this technical survey) to review the development of control, computing system design, communication, high-definition map (HD map), testing, and human behaviors in IVs. In addition, the third part (Part II for this technical survey) is to review the perception and planning sections. The objective of this article is to involve all the sections of AD, summarize the latest technical milestones, and guide abecedarians to quickly understand the development of AD and IVs. Combining the SoS and Part II, we anticipate that this work will bring novel and diverse insights to researchers and abecedarians, and serve as a bridge between past and future.

To address the coordination issue of connected automated vehicles (CAVs) at urban scenarios, a game-theoretic decision-making framework is proposed that can advance social benefits, including the traffic system efficiency and safety, as well as the benefits of individual users. Under the proposed decision-making framework, in this work, a representative urban driving scenario, i.e. the unsignalized intersection, is investigated. Once the vehicle enters the focused zone, it will interact with other CAVs and make collaborative decisions. To evaluate the safety risk of surrounding vehicles and reduce the complexity of the decision-making algorithm, the driving risk assessment algorithm is designed with a Gaussian potential field approach. The decision-making cost function is constructed by considering the driving safety and passing efficiency of CAVs. Additionally, decision-making constraints are designed and include safety, comfort, efficiency, control and stability. Based on the cost function and constraints, the fuzzy coalitional game approach is applied to the decision-making issue of CAVs at unsignalized intersections. Two types of fuzzy coalitions are constructed that reflect both individual and social benefits. The benefit allocation in the two types of fuzzy coalitions is associated with the driving aggressiveness of CAVs. Finally, the effectiveness and feasibility of the proposed decision-making framework are verified with three test cases.

To address the safety and efficiency issues of vehicles at multi-lane merging zones, a cooperative decision-making framework is designed for connected automated vehicles (CAVs) using a coalitional game approach. Firstly, a motion prediction module is established based on the simplified single-track vehicle model for enhancing the accuracy and reliability of the decision-making algorithm. Then, the cost function and constraints of the decision making are designed considering multiple performance indexes, i.e. the safety, comfort and efficiency. Besides, in order to realize human-like and personalized smart mobility, different driving characteristics are considered and embedded in the modeling process. Furthermore, four typical coalition models are defined for CAVS at the scenario of a multi-lane merging zone. Then, the coalitional game approach is formulated with model predictive control (MPC) to deal with decision making of CAVs at the defined scenario. Finally, testings are carried out in two cases considering different driving characteristics to evaluate the performance of the developed approach. The testing results show that the proposed coalitional game based method is able to make reasonable decisions and adapt to different driving characteristics for CAVs at the multi-lane merging zone. It guarantees the safety and efficiency of CAVs at the complex dynamic traffic condition, and simultaneously accommodates the objectives of individual vehicles, demonstrating the feasibility and effectiveness of the proposed approach.

This paper investigates a human-machine cooperative trajectory planning and tracking control approach for automated vehicles. The proposed method is developed based on a novel algorithm of cooperative human-machine rapidly-exploring random (HM-RRT) for path planning, together with the risk assessment of driver behavior. First, the driver’s behaviour is assessed according to the information of the predicted vehicle trajectory, the identified safe driving area and the driving risks evaluated in both lateral and longitudinal directions. Based on the driver’s expected driving task, when driving risks are identified by real-time assessment, then the human-machine cooperation is activated during trajectory planning. By HM-RRT, the newly developed safety assurance mechanism for path planning, the cooperative trajectory is then generated, which incorporates the driver’s desire and actions and automation’s corrective actions, to ensure the safety, stability and smoothness of the human-vehicle system. The simulation and experimental results show that the proposed HM-RRT algorithm can effectively improve the convergence rate and reduce the computation load, comparing to the conventional method. Beyond this, the proposed human-machine cooperation approach is able to simultaneously ensure the safety, stability and smoothness of the vehicle and largely reduce human-machine conflicts in real-time applications, demonstrating its feasibility and effectiveness.

Drones are a promising tool for parcel delivery, since they are cost-efficient and environmentally friendly. However, owing to the limited capacity of the on-board battery, their flight range is constrained. Thus, they cannot deliver some parcels if the customers are too far from the depot. To address this issue, this article proposes a novel method, in which a parcel delivery drone can “take” a public transportation vehicle and travel on its roof. The problem under consideration is how to make use of the public transportation network to route the drone between the depot and the customer. Compared to the currently available methods that use drones, the most important merit of this approach is a significant expansion of the delivery area. We construct a multimodal network consisting of public transportation vehicles' trips and drone flights. Because of the complexity of this multimodal network, we convert it to a simple network with a set of simple procedures. In the extended network, we formulate the shortest drone path problem that minimizes the return instant to the depot, subject to that the drone energy consumption on this path is no greater than the initial energy. We present a Dijkstra-based method to find the shortest drone path. Moreover, we extend the proposed method to the case with uncertainty, because the public transportation vehicles cannot exactly follow their timetables in practice. Simulation results are presented to demonstrate how the method works.

Considering that human-driven vehicles and autonomous vehicles (AVs) will coexist on roads in the future for a long time, how to merge AVs into human drivers' traffic ecology and minimize the effect of AVs and their misfit with human drivers, are issues worthy of consideration. Moreover, different passengers have different needs for AVs, thus, how to provide personalized choices for different passengers is another issue for AVs. Therefore, a human-like decision making framework is designed for AVs in this paper. Different driving styles and social interaction characteristics are formulated for AVs regarding driving safety, ride comfort and travel efficiency, which are considered in the modeling process of decision making. Then, Nash equilibrium and Stackelberg game theory are applied to the noncooperative decision making. In addition, potential field method and model predictive control (MPC) are combined to deal with the motion prediction and planning for AVs, which provides predicted motion information for the decision-making module. Finally, two typical testing scenarios of lane change, i.e., merging and overtaking, are carried out to evaluate the feasibility and effectiveness of the proposed decision-making framework considering different human-like behaviors. Testing results indicate that both the two game theoretic approaches can provide reasonable human-like decision making for AVs. Compared with the Nash equilibrium approach, under the normal driving style, the cost value of decision making using the Stackelberg game theoretic approach is reduced by over 20%.

Drones have been regarded as a promising means for future delivery industry by many logistics companies. Several drone-based delivery systems have been proposed but they generally have a drawback in delivering customers locating far from warehouses. This paper proposes an alternative system based on a public transportation network. This system has the merit of enlarging the delivery range. As the public transportation network is actually a stochastic time-dependent network, we focus on the reliable drone path planning problem (RDPP). We present a stochastic model to characterize the path traversal time and develop a label setting algorithm to construct the reliable drone path. Furthermore, we consider the limited battery lifetime of the drone to determine whether a path is feasible, and we account this as a constraint in the optimization model. To accommodate the feasibility, the developed label setting algorithm is extended by adding a simple operation. The complexity of the developed algorithm is analyzed and how it works is demonstrated via a case study.

Connected automated driving technologies have shown tremendous improvement in recent years. However, it is still not clear how driving behaviors and energy consumption correlate with each other and to what extent these factors related to connected vehicles can influence the motion prediction performance. The precise recognition of driving behaviors and prediction of the vehicle motion is critical to the driving safety for connected automated vehicles (CAVs). Hence, in this study, an energy-aware driving pattern analysis and motion prediction system are proposed for CAVs using a deep learning-based time-series modeling approach. First, energy-aware longitudinal acceleration and deceleration behaviors and lateral lane-change behaviors are statistically analyzed. Then, a sliding standard deviation (SSD) test is applied to evaluate the smoothness of the trajectory and velocity signals considering different energy consumption levels. An energy-aware personalized joint time-series modeling (PJTSM) approach based on a deep recurrent neural network (RNN) and long short-term memory (LSTM) cell are proposed for accurate motion (trajectory and velocity) prediction of the leading vehicle. Finally, the differences in the prediction performance regarding different energy consumption levels are compared and discussed. It is shown that due to the higher randomness of the driving behaviors, the prediction accuracy for heavy energy users is the lowest among the three categories, which means it is harder to anticipate the driving behaviors of cars exhibiting heavy energy consumption. The personalized estimation of driving behaviors of CAVs will contribute to safer automated driving and transportation systems.

The automated driving vehicles are experiencing a rapid development in worldwide recently. It is commonly believed that before the achievement of fully autonomous driving, the driver will always need to remain within the vehicle control loop. Hence, intelligent interaction and collaboration between the human driver and the automation will be an efficient solution for the improvement of road safety, traffic efficiency, and social acceptance to the automated vehicles. As a popular collaboration method, shared control has been widely studied in the past two decades. While it is still a challenging task to involve rich human factors into the shared control system to increase the driving experience and acceptance of the automation. In this study, a literature review on human-centered shared control is proposed towards solid research on driver-vehicle collaboration. First, the basic background and literature surveys on the human-machine collaboration (HMC) is proposed, and the important factors for efficient multi-agent collaboration and teaming are discussed. Then, different driver behavior and state modeling methods are reviewed. Based on the HMC schemes and driver behavior recognition techniques, literature surveys on human-centered shared control are proposed. Finally, challenges and future works on human-centered shared control are analyzed.

In this paper, a novel reference-free shared control system is designed for obstacle avoidance for automated vehicles. Rather than using a reference path to guide the driver, the proposed framework constrains the vehicle's status to guarantee the safety without scarifying the driver's freedom. The constrained Delaunay triangle method is introduced to identify the vehicle's position constraints and the constraints of obstacle avoidance, vehicle stability and physical limitations are investigated and unified. A nonlinear predictive control problem, which is constructed accounting nonlinear vehicle dynamics and given driver actions, is designed to optimize the steering and braking actions needed to keep the vehicle safe. The automation is supposed to correct the driver's steering or braking actions to prevent constraint violation and losing the control of vehicle. The simulation results show that the automation can assist the driver to avoid obstacles and guarantee the vehicle's stability with minimal control intervention.

In a Steer-by-Wire (SbW) system, the mechanical linkages which connect the steering wheel to front wheels are replaced by a digitally controlled steering system. In spite of improved vehicle safety due to better steering capability, the SbW system actuator failure may lead to unwanted steering performance or even instability. A fault tolerant model predictive control (MPC) with fault compensation for SbW systems based on delta operator for actuator faults is proposed. It deploys an observer to estimate both the fault information and the faulty SbW system states. At each sampling time, the MPC immediately compensates for the fault. The gains of the observer and the fault tolerant MPC controller are obtained by solving a linear matrix inequality derived from the Lyapunov theory. The simulation results illustrate that the proposed fault tolerant control strategy can counter the various types of actuator faults and maintain superior steering performance to shift operator-based MPC controller.

A parcel delivery system with a unmanned aerial vehicle (UAV) and a public train is presented, where the train moves naturally as it is and the UAV can depart the train to deliver parcels. The UAV can travel with the train and replace its battery on the roof. An optimization problem is formulated to minimize the total delivery time and two algorithms are proposed. The exact algorithm gives the optimal schedule, but it is not scalable. The developed suboptimal algorithm is computationally efficient and achieves close performance to that of the exact algorithm. Realistic simulations are conducted to evaluate the proposed algorithms and they are compared with existing schemes.

Though drones have become a promising tool for parcel delivery, due to the limited capacity of the on-board battery, their flight range is constrained. To enlarge the coverage range, the paper proposes a novel method, in which a parcel delivery drone can "take" a public transportation vehicle and travel on its roof. We investigate how to make use of the public transportation network to route the drone between the depot and a customer. Considering the complexity of the public transportation network, its randomness and time-dependency, instead of off-line planning a global optimal path, an adaptive algorithm is developed. Simulation results are presented to demonstrate how the proposed approach works.

The Steer-by-Wire (SbW) system is an electronically controlled steering system that is able to improve steering capability without mechanical links between the steering wheel and the front wheels. However, failure of the SbW system actuator may lead to steering performance degradation and result in instability. In this paper, a fault tolerant sliding mode predictive control (SMPC) strategy for an SbW system is proposed. The sliding mode control is applied to improve the robustness of the model predictive control (MPC) in the presence of modeling uncertainties and disturbances, while the MPC is applied to enhance the fault tolerant capability of the steering control processes. The chaos particle swarm optimization (CPSO) algorithm is introduced to optimize the MPC and a two-stage Kalman filter is introduced to simultaneously provide fault information and state estimation. The performance of the proposed approach is validated through computer simulation. The results demonstrate that the proposed SMPC-CPSO controller is more robust and provides better tracking performance in the presence of model uncertainties, disturbance, and actuator faults than SCMP-PSOs (heterogeneous comprehensive learning particle swarm optimization, evolutionary particle swarm optimizer, etc), SMPC-differential evolution, MPC, SMPC, and MPC-PSO.

This paper considers the data delivery delay problem in wireless sensor networks. The delivery delay is a significant measure when the data freshness is the first concern of users. The goal of this paper is to route delay-sensitive data to mobile nodes (M nodes) within an allowed latency. The data collection system is composed of a set of M nodes amounted on buses, a set of sensor nodes to detect the interested phenomenon and a set of special nodes deployed at bus stops to assist data routing. An optimization-based approach called improved-unusual message delivery path construction (I-UMDPC) is proposed. Considering the actual features of bus operation, two aspects of uncertainties are accounted in our approach: 1) the bus arrival time and 2) the stop duration. Extensive simulations as well as practical experiments on our testbed demonstrate that I-UMDPC is able to route delay-sensitive data reliably and efficiently and performs better than existing work.