Agile signal switching under frequent arrivals of transit vehicles, combined with the need to respect multiple operational constraints, presents significant challenges for effective and safe signal control, as well as for the real-world implementation of reinforcement learning-based control algorithms. We introduce a safe reinforcement learning-based fully adaptive multimodal traffic signal controller in a connected vehicle environment that incorporates a cost estimator during the learning process to account for multiple operational constraints. It utilises the Duelling Double Deep Q -network and a multicriteria reward to minimise passenger delay, and maximise throughput under lower and upper bounds on green time and maximum phase skip constraint. Unsafe situations due to inappropriate and frequent phase switches are specified as a safety constraint, which constrains the learning process. The Lagrangian method is used to transform the constrained learning to an unconstrained one based on the concept of safe reinforcement learning, and the associated Lagrange multiplier is updated via a gradient-based mechanism. The performance of the proposed algorithm is evaluated for an isolated intersection using simulations in SUMO under different traffic demands, fixed public transport schedules and random passenger occupancy levels. The results demonstrate that the proposed algorithm reduces queue length and public transport passenger delays compared to state-of-the-art model-based and model-free signal controllers. The integration of a cost estimator effectively handles both hard and soft constraints during learning. The proposed algorithm resolves public transport priority request conflicts, makes a trade-off between public transport and individual traffic, and ensures traffic safety.

Network signal control is an effective way to mitigate traffic congestion. However, most network signal control methods ignore the risk of queue spillback. Although local and decentralized control methods have the potential to address spillback, their performance at the system/network level is not guaranteed, making it challenging to achieve the global optimum. This study proposes a novel hierarchical model predictive control (MPC) approach that utilizes the link transmission model (LTM) with queue transmission at both road segment and turn levels to optimize traffic signals for road networks. At the network level, the controller employs a state-of-the-art LTM framework that can describe segment-level flow dynamics and uses quadratic programming to determine the effective fractions of green time for network throughput maximization. The network decisions of green time fractions are sent to the local layer as a reference. The local layer builds on a refined LTM with turn-level queue transmission and formulates a nonlinear programming problem to track the reference to ensure that the optimal network decision is realized at the individual intersections while eliminating potential spillback. Simulation experiments on both an arterial and a grid network are conducted to verify the performance of the proposed approach. The results reveal that the proposed MPC leads to substantial improvements in terms of throughput and queue length, especially in oversaturated conditions, and is robust against demand prediction errors.

The deep understanding of the behaviors of traffic participants is essential to guarantee the safety of automated vehicles (AV) in mixed traffic with vulnerable road users (VRUs). Precise trajectory prediction of traffic participants can provide reasonable solution space for motion planning of AV. Early works mainly focused on handcrafting the feature representation and designing complicated architectures in deep learning-based prediction models. However, these approaches overlooked the fact that different road users perceive the safety of the same interaction differently and also exhibit heterogeneous risk-taking styles. In this paper, we will develop a model for trajectory prediction based on risk-taking styles. The model accounts for the expected positions and occupancy of traffic participants in the surrounding environment. It consists of two sequential steps: risk-taking styles of multi-modal road users under interactive scenes are first clustered, and then reformulated in the heterogeneous graph model for trajectory prediction. The model is validated by the driving data collected on the urban road using a public dataset. Comparative experiments demonstrate that the proposed method can predict the trajectory of traffic participants much more accurately than the state-of-the-art methods.

Interacting with human road users is one of the most challenging tasks for autonomous vehicles. For congruent driving behaviors, it is essential to recognize and comprehend sociality, encompassing both implicit social norms and individualized social preferences of human drivers. To understand and quantify the complex sociality in driving interactions, we propose a Virtual-Game-based Interaction Model (VGIM) that is parameterized by a social preference measurement, Interaction Preference Value (IPV). The IPV is designed to capture the driver’s relative inclination towards individual rewards over group rewards. A method for identifying IPV from observed driving trajectory is also developed, with which we assessed human drivers’ IPV using driving data recorded in a typical interactive driving scenario, the unprotected left turn. Our findings reveal that (1) human drivers exhibit particular social preference patterns while undertaking specific tasks, such as turning left or proceeding straight; (2) competitive actions could be strategically conducted by human drivers in order to coordinate with others. Finally, we discuss the potential of learning sociality-aware navigation from human demonstrations by incorporating a rule-based humanlike IPV expressing strategy into VGIM and optimization-based motion planners. Simulation experiments demonstrate that (1) IPV identification improves the motion prediction performance in interactive driving scenarios and (2) the dynamic IPV expressing strategy extracted from human driving data makes it possible to reproduce humanlike coordination patterns in the driving interaction.

Interacting with other vehicles while ensuring safety is a routine task for human drivers, but it can pose a challenge for autonomous vehicles. To address this challenge, we derived a generic utility term of interaction activeness (UTIA) from the driving interaction formulation which considers the rationality of interacting counterparts. Our research shows that incorporating UTIA as a supplementary utility term can improve the active interaction capability of both sampling-based and game-theoretic baseline motion planners without compromising safety. Through simulation experiments, we observed that on average, incorporating the weighted UTIA into the utility function of baseline planners can result in an 8.8% increase in the success rate of exiting a highway within a set distance.

Path dispersion (the spatial distribution of vehicular paths) is an important feature of traffic flow inside intersections and differs from traffic flow running along traffic lanes at road segment, especially under conflicting movements. The path dispersion reflects the operational features of traffic flow and is related to driving behaviour, arrival flow patterns, layout design, and the traffic control and management scheme. This study aims to improve the understanding of the overall path dispersion of unprotected left-turns and the opposing through movement. A behavioural simulation model was established to represent the overall path dispersion. Human behaviours regarding vehicle trajectory planning with and without conflicting vehicles were modelled based on optimal control and integrated into the proposed discrete event simulation framework. The descriptive power and accuracy of the proposed simulation model were validated using empirical data. The effects of the spatial size of the intersection, crossing angle, and traffic volume on the path dispersion of the left-turn and through movement were explored based on numerical experiments. The results show that the proposed simulation model can represent the path dispersion of left-turn and opposing through movement well for both the calibrated intersections and newly added intersections without model parameter recalibration with an average error of 8.92%.

Real-time safety systems are crucial components of intelligent vehicles. This paper introduces a prediction-based collision risk assessment approach on highways. Given a point mass vehicle dynamics system, a stochastic forward reachable set considering two-dimensional motion with vehicle state probability distributions is firstly established. We then develop an acceleration prediction model, which provides multi-modal probabilistic acceleration distributions to propagate vehicle states. The collision probability is calculated by summing up the probabilities of the states where two vehicles spatially overlap. Simulation results show that the prediction model has superior performance in terms of vehicle motion position errors, and the proposed collision detection approach is agile and effective to identify the collision in cut-in crash events.

We present a simulation-based approach to assess the safety impacts of vehicles equipped with Automated Driving Systems (ADS) in mixed traffic with Human-driven Vehicles (HV). Specifically, we compare two generic longitudinal strategies of ADS to handle a cut-in: Reactive ADS acting only when the cut-in vehicle crosses the target lane boundary, and Predictive ADS acting at the onset of the cut-in manoeuvre. We identify their distinctive effects on the traffic safety under cut-in maneuvers of adjacent human-driven vehicles at highway merges. We employ a microscopic traffic flow simulator that describes the lane changing process with high detail, accounting for the vehicle interaction and consequent trajectory updates. These high-resolution trajectories are post-processed to estimate a set of relevant surrogate measures of safety. By analyzing these measures, we find that the predictive ADS significantly outperforms the reactive ADS in aspects such as temporal proximity to crash, expected crash severity and the driving risk (combining the two aspects), and the number of aborted lane changes by HV. The negative safety impact of reactive ADS becomes prominent at penetration rate > 10%. The major difference between the two ADS approaches appears in the dynamics of risk during the lane changing. When a vehicle cuts in ahead of Reactive ADS, the risk peaks approximately halfway through the maneuver; whereas with Predictive ADS the risk remains marginal throughout. This work demonstrates the potential of simulation-based safety assessment to differentiate the safety impacts of automation functionalities at an early stage of product development.

Connected Automated Vehicles (CAVs) have the potential to improve traffic operations when they cooperatively maneuver in merging sections. State-of-the-art approaches in cooperative merging either build on heuristics solutions or prohibit mainline CAVs to change lane on multilane highways. This paper proposes a hierarchical cooperative merging control approach that ensures collision-free and traffic-efficient merging through the interaction of a maneuver planner and an operational trajectory controller. The planner predicts future vehicular trajectories, including acceleration trajectories and time instants when lane changes start, in a long horizon up to 50 seconds with a linear prediction model. It establishes the optimal dynamic vehicle sequence in each lane by minimizing predicted traffic disturbances that can propagate upstream and lead to traffic breakdown. During the process, mainline vehicles may change lane to facilitate the on-ramp merging, albeit with a higher ego cost. The operational controller follows the established instructions from the planner and regulates vehicular trajectories with model predictive control in a shorter horizon of 6 seconds. The performance of the designed hierarchical cooperative merging control approach was compared to a cooperative merging method utilizing widely used first-in-first-out rule to establish merging sequences and the same operational controller to generate vehicular trajectories. Systematic comparison shows that the proposed approach consistently results in less disturbances during merging under 528 different scenarios with different traffic states, initial vehicular states, and desired time gap settings. On average, a decrease of 39.18% in disturbances was observed.

Road traffic safety has attracted increasing research attention, in particular in the current transition from human-driven vehicles to autonomous vehicles. Surrogate measures of safety are widely used to assess traffic safety but they typically ignore motion uncertainties and are inflexible in dealing with two-dimensional motion. Meanwhile, learning-based lane-change and trajectory prediction models have shown potential to provide accurate prediction results. We therefore propose a prediction-based driving risk metric for two-dimensional motion on multi-lane highways, expressed by the maximum risk value over different time instants within a prediction horizon. At each time instant, the risk of the vehicle is estimated as the sum of weighted risks over each mode in a finite set of lane-change maneuver possibilities. Under each maneuver mode, the risk is calculated as the product of three factors: lane-change maneuver mode probability, collision probability and expected crash severity. The three factors are estimated leveraging two-stage multi-modal trajectory predictions for surrounding vehicles: first a lane-change intention prediction module is invoked to provide lane-change maneuver mode possibilities, and then the mode possibilities are used as partial input for a multi-modal trajectory prediction module. Working with the empirical trajectory dataset highD and simulated highway scenarios, the proposed two-stage model achieves superior performance compared to a state-of-the-art prediction model. The proposed risk metric is computationally efficient for real-time applications, and effective to identify potential crashes earlier thanks to the employed prediction model.

Current lane-based microscopic traffic simulators combine car-following and lane changing logic to describe the (often discrete) lateral vehicle motion on multi-lane road segments. However, the simulated lateral trajectories are physically unplausible and inside-lane behavior such as lane-keeping and curve negotiation cannot be modelled. In this work, we integrate lateral vehicle dynamics and yaw motion into a traffic simulation framework, aiming to describe lateral motion and vehicle interactions with more precision. The resulting framework consists of two coupled layers, an upper tactical level that plans maneuvers such as lane-changing; and a lower operational layer with a control module (steering and acceleration control) that operates in a closed loop with the bicycle model of vehicle dynamics. The feedback mechanism between the layers allows for dynamic trajectory re-planning. Unlike the microscopic traffic models, the proposed framework accounts for lateral vehicle dynamics and yaw motion; provides additional variables such as vehicle heading and front wheel steering angle; and is hence termed as submicroscopic. Case study results demonstrate the power of the framework to include lateral maneuvers such as curve negotiation, corrective steering, lane change abortion and fragmented lane changing. The framework was operationalized to model multi-lane traffic flow consisting of human-driven vehicles. At the macroscopic level, the traffic flow simulation can reproduce phenomena such as capacity drop. Thus the framework preserves the properties of the component models and at the same time describe the continuous 2-D planar movement of vehicles.

Gap selection and dynamic speed profiles of interacting vehicles at on-ramps affect the safety and efficiency of highway merging sections. This paper puts forward a hierarchical control approach for Connected Automated Vehicles (CAVs) to achieve efficient and safe merging operations. A tactical layer controller employs a second-order car-following model with a cooperative merging mode to represent a cooperative merging process and generates an optimal vehicle merging sequence and time instants when on-ramp CAVs start to adapt their speeds and positions to prepare merging into the target gaps respectively. An operational layer controller is designed based on Model Predictive Control (MPC). It uses a third-order vehicle dynamics model and optimizes desired accelerations for CAVs and the time instants when the on-ramp CAVs initiate the lane-changing executions respectively. Both the tactical layer controller and operational layer controller derive their control commands by minimizing an objective function for different time horizons. The objective function penalizes deviations of CAVs’ inter-vehicle gaps to their desired values, relative speeds to their direct predecessors, and actual or desired accelerations, subject to constraints on velocities, actual or desired accelerations, and inter-vehicle gaps. The performance of the proposed hierarchical control framework and a benchmark on-ramp merging method using afirst-in-first-outrule to determine the merging sequence is demonstrated under 135 scenarios with different initial conditions, desired time gap settings, and numbers of on-ramp vehicles. The experimental results show the superiority of the hierarchical control approach.

In this research, an optimal control-based Cooperative Adaptive Cruise Control (CACC) system is proposed. The proposed system is able to enforce a target time gap between platoon members and is formulated in the space domain instead of the time domain which is adopted by most optimal control-based CACC systems in the past. By having this change, its robustness against communication failure is greatly improved and thus minimum safety headway buffer is reduced which leads to better mobility. In addition, third-order vehicle dynamics are modeled into the proposed control in order to improve control precision when implemented in the field. Local stability and string stability are theoretically proven. The proposed system is evaluated by simulation. Results reveal that the proposed CACC system outperforms the state-of-the-artH∞synthesis-based controller and linear feedback-based controller. The benefit of fuel consumption reduction ranges from 0.35% to 16.11%, while the benefit of CO2emission ranges from 0.48% to 12.40%. Furthermore, the proposed CACC improves local stability from 11.03% to 25.90%, and string stability by up to 23.82%. The computation speed of the proposed method is 1.26 ms (with prediction horizon as 1.5 s and resolution as 0.1 s) on a regular laptop which indicates the proposed system's potential to be applied in real-time.

Cooperative Adaptive Cruise Control (CACC) in previous researches typically refers to the linear controller with a gap policy. The system could not be designed to fulfill multiple objectives. This inspires the concept of optimal control based CACC in this paper. The basic procedure of the proposed controller is to gather the information collected by each vehicle to the computation unit first, then plan the trajectory of all the followers by solving an optimal control problem, and dispatch the optimal motion command to each vehicle at last. This paper models CACC under optimal control framework. A numerical approach inspired by dynamic programming is adopted to solve the control problem. The stability of the proposed controller is thoroughly investigated in terms of both local stability and string stability. To verify the concept of controller, solution, and the analysis about stability, simulation is carried out. The simulation verifies that the numerical method is effective with respect to computation time. Both theoretical analysis and simulation proved that the proposed optimal control based CACC is both local stable and string stable. The low computation burden, local stability, and string stability together guarantee the future implementation of the proposed controller.

Arterial traffic signal coordination makes traffic flow more efficient and safer. This paper presents a partition-enabled multi-mode band (PM-BAND) model that is designed to solve the signal coordination problem for arterials with multiple modes, i.e., passenger cars and transit vehicles. The proposed method permits the progression bands to be broken if necessary and optimizes system partition and signal coordination in one unified framework. The impacts of traffic demand of passenger cars and transit vehicles as well as the geometry characteristics of the arterials are taken into account. Signal timings and waiting time of transit vehicles at stations are optimized simultaneously. The PM-BAND model is formulated as a mixed-integer linear program, which can be solved by the standard branch-and-bound technique. Numerical example results have demonstrated that the PM-BAND model can significantly reduce the average number of stops and delay compared with the other models, i.e., MAXBAND and MULTIBAND. Moreover, the progression bands generated by the PM-BAND model have a higher reliability and effectiveness.

We design controllers and derive implementable algorithms for autonomous and cooperative car-following control (CFC) systems under a receding horizon control framework. An autonomous CFC system controls vehicle acceleration to optimize its own situation, whereas a cooperative CFC (C-CFC) system coordinates accelerations of cooperative vehicles to optimize the joint situation. To realize simultaneous control of many vehicles in a traffic system, decentralized and distributed algorithms are implemented in a microscopic traffic simulator for CFC and C-CFC controllers, respectively. The impacts of the proposed controllers on dynamic traffic flow features, particularly on formation and propagation of moving jams, are investigated through a simulation on a two-lane freeway with CFC/C-CFC vehicles randomly distributed. The simulation shows that the proposed decentralized CFC and distributed C-CFC algorithms are implementable in microscopic simulations, and the assessment reveals that CFC and C-CFC systems change moving jam characteristics substantially.