Interacting with diverse and stochastic traffic participants is a critical challenge for autonomous vehicles (AVs), as it necessitates advanced decision-making systems to replicate the natural adaptability of human drivers. In particular, navigating safely and efficiently in dense traffic scenarios poses a significant challenge for decision-making, which is inherently an interactive task,i.e., nearby traffic participants will influence AVs’ action, and vice versa. Decision-making solutions that rely solely on unidirectional interaction schemes, neglecting the mutual influence between AVs and other traffic participants, may lead to overly defensive behaviors or the “freezing robot problem”. In recent years, researchers have been increasingly focused on incorporating bidirectional interactions into the decision-making process to make safe, intelligent, and socially compatible decisions. Currently, a comprehensive review of interaction-aware decision-making techniques remains lacking. To this end, this paper aims to provide a systematic review of interaction-aware decision-making methodologies for autonomous driving. Specifically, this paper analyzes the challenges in considering bidirectional interactions between AVs and other traffic participants. In addition, the state-of-the-art techniques for interaction-aware decision-making solutions are reviewed. More importantly, simulation and benchmarks for interaction-aware decision-making validation are also presented. Finally, research perspectives are highlighted to facilitate future studies for interaction-aware decision-making policy design.

Driving on public roads is inherently an interactive task, i.e., autonomous vehicles’ (AVs) actions will influence nearby traffic participants’ reactions, and vice versa. Decision-making for AVs in highly interactive driving scenarios (e.g., dense traffic) requires accurately forecasting the impact of the AVs’ intention on nearby traffic participants’ motion. To this end, a hierarchical decision-making framework (HDM) is proposed to navigate through interactive scenarios safely and efficiently. Specifically, the upper layer of the HDM serves as a coarse-level policy generator, which utilizes the plan-informed graph attention network (P-GAT) to provide interaction-aware guidance. The P-GAT predictor takes the historical states of nearby traffic participants, road structure information, and AVs’ potential intentions as inputs. Subsequently, it predicts the motions of other traffic participants in response to potential actions of the AVs, which is then systematically evaluated to generate interactive guidance. Furthermore, the learned policy is utilized to guide the lower layer, which utilizes a fine-level model predictive control (MPC)-based planner to ensure safety and kinematic feasibility. Finally, to validate the effectiveness of HDM, both qualitative and quantitative experiments are carried out. More importantly, the hardware-in-the-loop (HiL) experiment is also implemented, including mandatory lane change in dense traffic flow and interaction with the human driver. The results demonstrate that the proposed HDM can improve driving safety and efficiency compared with baselines.

Autonomous driving confronts great challenges in complex traffic scenarios, where the SOTIF risk can be triggered by the dynamic operational environment and system insufficiencies. The SOTIF risk is reflected not only intuitively in the collision risk with objects outside the autonomous vehicles, but also inherently in the performance limitation risk of the implemented algorithms. How to minimize the SOTIF risk for autonomous driving is currently a critical, difficult, and unresolved issue. Therefore, this paper proposes the “Self-Surveillance and Self-Adaption System” as a systematic approach to online minimize the SOTIF risk, which aims to provide a systematic solution for monitoring, quantification, and mitigation of inherent and external risks. As a demonstration of the system, the risk monitoring of the perception algorithm is highlighted. Moreover, the inherent perception algorithm risk and external collision risk are jointly quantified via SOTIF entropy, which is then propagated downstream to the decision-making module and mitigated. Finally, Hardware-in-the-Loop experiments are conducted to verify the efficiency and effectiveness of the system. The results demonstrate that the system enables dependable online monitoring, quantification, and mitigation of SOTIF risk in real-time critical traffic environments.

Reinforcement learning (RL) has been widely used in the decision-making of autonomous vehicles (AVs) in recent studies. However, existing RL methods generally find the optimal policy by maximizing the expectation of future returns, which lacks distributional treatments of risky situations. Additionally, various uncertainties arising from the environment could also cause unreliable decisions, particularly in some complex urban environments. In this paper, the fully parameterized quantile network (FPQN) is utilized to estimate the full return distribution. Then, the conditional value-at-risk (CVaR) is utilized with the return distribution information to generate uncertainty-aware driving behavior. Additionally, an uncontrolled four-way intersection is developed by the Simulation of Urban Mobility (SUMO) simulation platform, which considers both the surrounding vehicles (SVs) and pedestrians. More specifically, to simulate the real-world traffic environment, the uncertainty arising from the occlusion, and the behavior uncertainty of surrounding traffic participants are also considered. The experiment results suggest that the proposed method outperforms the baseline methods in terms of safety. Furthermore, the results also indicate that the proposed method can make reasonable decisions in some challenging driving cases in the presence of uncertainty.

Motion prediction modules are crucial for autonomous vehicles to forecast the future behavior of surrounding road users. Failures in prediction modules can mislead a downstream planner to make unsafe decisions. Currently, deep learning technology has been widely used to design prediction models due to its impressive performance. However, such models may fail in long-tail driving scenarios where the training data are insufficient or unavailable, which represents the so-called epistemic uncertainty of prediction models. This paper proposes a risk-aware decision-making (RADM) framework to handle the epistemic uncertainty arising from training the prediction model on insufficient data. First, a multi-agent prediction network with epistemic uncertainty quantification is proposed. This network uses the historical states of nearby road users, map information, and traffic lights as inputs. Then, the RADM utilizes model predictive control technique to not only process the multi-agent prediction results but also to consider the epistemic uncertainty of the prediction model. In addition, the accuracy of the established prediction model is verified on real-world driving datasets. Furthermore, the proposed RADM is evaluated on the log-replay data obtained from real-world driving logs and using the SUMO simulator, considering multiple challenging cases where pedestrians and non-motorized vehicles cross the intersection illegally. The experimental results demonstrate that RADM can reduce the driving risk and improve driving safety and supplementary videos are provided athttps://github.com/SOTIF-AVLab/RADM.

Autonomous driving is considered one of the revolutionary technologies shaping humanity’s future mobility and quality of life. However, safety remains a critical hurdle in the way of commercialization and widespread deployment of autonomous vehicles on public roads. Safety concerns require the autonomous driving system to handle uncertainties from multiple sources that are either preexisting, e.g., the stochastic behavior of traffic participants or scenario occlusion, or introduced as a result of processing, e.g., the application of neural networks. Thus, it is crucial to analyze the sources of uncertainties and quantify the risks associated with them, including the propagated risks that accumulate in the decision-making system. In this context, this paper provides an overview of uncertainty challenges and state-of-the-art techniques for mitigating these challenges. We argue that the uncertainties mainly originate from two aspects: 1) the external traffic environment, and 2) the internal autonomous driving system. Specifically, this paper first analyzes the safety challenges caused by the uncertainties and summarizes their sources. In addition, the corresponding techniques that mitigate and quantify the risk of uncertainties are presented. Finally, research perspectives are highlighted to facilitate future studies for guaranteeing the safety of autonomous vehicles.