Advances in Uncertainty-aware Intelligent Driving

The uncertainty-aware Intelligent Driving is an emerging field that addresses the challenges of intelligent driving vehicles in complex and unpredictable environments. The concept of uncertainty in this context refers to the variability and unpredictability of road users’ behavior, as well as the limitations of sensors and computational capabilities. Such uncertainties pose great challenges to the driving tasks of intelligent vehicles, which should be able to perceive and interact with all road users, localize and navigate through dense traffic, and make real-time predictions and decisions based on complex environmental information, incomplete or lost data, noise influence, and even adversarial attacks.

To address the challenges of uncertainties in intelligent driving, researchers attempt to develop algorithms and systems that can account for uncertainty in various related tasks. It requires a comprehensive understanding of the driving environment, including the behavior of other road users, traffic rules, and road conditions. Thus, the perception, prediction, planning, and control algorithms of intelligent vehicles are continuously improved to be more capable of handling complex and uncertain situations. Common techniques to estimate and manage uncertainty are probabilistic models, machine learning, sensor fusion, and optimization strategies. By incorporating these uncertainty-aware techniques, the safety, efficiency, and robustness of intelligent driving can be further improved in real-world driving scenarios.

The aim of this Research Topic is to address these challenges by presenting the related research advances in the hope of promoting uncertainty-aware intelligent driving techniques. In this Research Topic, original research articles and reviews are welcome. Topics may include (but are not limited to) the following:
- Insights on the challenges in uncertainties in intelligent driving, their causes, and related factors;
- Improvements in object detection, mapping and localization, trajectory prediction, and behavior recognition of road traffic participants in complex environments;
- Novel planning, decision, and control approaches to account for uncertainties in intelligent driving-related scenarios;
- New defense methods against natural or manual adversarial attacks and the related analysis;
- Protection strategies for drivers or passengers related to driving uncertainties;
- Simulation and evaluation tool related to intelligent driving uncertainties.

Keywords: Intelligent Driving, Environmental Perception, Mapping and Localization, Trajectory Prediction, Behavior Decision, Adversarial Defense

Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.