Tracking and behavior reasoning of surrounding vehicles on a roadway are keys for the development of automated vehicles and an advanced driver assistance system (ADAS). Based on dynamic information of the surrounding vehicles from the tracking algorithm and driver intentions from the behavior reasoning algorithm, the automated vehicles and ADAS can predict possible collisions and generate safe motion to avoid accidents. This paper presents a unified vehicle tracking and behavior reasoning algorithm that can simultaneously estimate the vehicle dynamic state and driver intentions. The multiple model filter based on various behavior models was used to classify the vehicle behavior and estimate the dynamic state of surrounding vehicles. In addition, roadway geometry constraints were applied to the unified vehicle tracking and behavior reasoning algorithm in order to improve the dynamic state estimation and the behavior classification performance. The curvilinear coordinate system was constructed based on the precise map information in order to apply the roadway geometry to the tracking and behavior reasoning algorithm. The proposed algorithm was verified and evaluated through experiments under various test scenarios. From the experimental results, we concluded that the presented tracking and behavior reasoning algorithm based on the roadway geometry constraints provides sufficient accuracy and reliability for automated vehicles and ADAS applications.

Detection and tracking of moving objects is one of the most essential functions of autonomous cars. In order to estimate the dynamic information of a moving object accurately, laser scanners are widely used for their highly accurate distance data. However, these data only represent the surface of an object facing the sensor and changes the appearance of an object over time. This change produces unexpected tracking errors of estimated dynamic states. In this paper, in order to minimize the tracking error caused by appearance changes, a tracking algorithm based on L-shaped model switching is proposed. The suggested algorithm is validated in real traffic experiments where position, velocity, and heading angle error were measured by using precise GPS. The L-shape tracking algorithm successfully mitigated the effect of appearance changes and improved estimation performance.

Predicting the trajectories of surrounding vehicles by considering their interactions is an essential ability for the functioning of autonomous vehicles. The subsequent movement of a vehicle is decided based on the multiple maneuvers of surrounding vehicles. Therefore, to predict the trajectories of surrounding vehicles, interactions among multiple maneuvers should be considered. Recent research has taken into account interactions that are difficult to express mathematically using data-driven deep learning methods. However, previous studies have only considered the interactions among observed trajectories due to subsequent maneuvers that are unobservable and numerous maneuver combinations. Thus, to consider the interaction among multiple maneuvers, this paper proposes a hierarchical graph neural network. The proposed hierarchical model approximately predicts the multiple maneuvers of vehicles and considers the interaction among the maneuvers by representing their relationships in a graph structure. The proposed method was evaluated using a publicly available dataset and a real driving dataset. Compared with previous methods, the results of the proposed method exhibited better prediction performance in highly interactive situations.

This paper presents track fusion and behavioral reasoning for moving vehicles in close proximity based on the curvilinear coordinates of roadway geometries. The inferred track and behavior of other vehicles can be used to perform safe actions in intelligent vehicle applications and autonomous driving. Vehicle detections from multiple perception sensors are integrated using track-to-track (T2T) fusion based on a cross-covariance method, and this T2T fusion is performed with curvilinear coordinates created using prebuilt roadway geometry on a digital map. The coordinate conversion to curvilinear space has many benefits for behavioral reasoning and tracking, such as constraining problem spaces and dimensions. A machine learning classifier based on a support vector machine is then applied to deduce the behavior of nearby vehicles. The algorithms presented here for track fusion and behavioral reasoning based on curvilinear coordinates have been verified through experiments in various real traffic scenarios.

Multi-sensor perception systems may have mismatched coordinates between each sensor even if the sensor coordinates are converted to a common coordinate. This discrepancy can be due to the sensor noise, deformation of the sensor mount, and other factors. These mismatched coordinates can seriously affect the estimation of a distant object’s position and this error can result in problems with object identification. To overcome these problems, numerous coordinate correction methods have been studied to minimize coordinate mismatching, such as off-line sensor error modeling and real-time error estimation methods. The first approach, off-line sensor error modeling, cannot cope with the occurrence of a mismatched coordinate in real-time. The second approach, using real-time error estimation methods, has high computational complexity due to the singular value decomposition. Therefore, we present a fast online coordinate correction method based on a reduced sensor position error model with dominant parameters and estimate the parameters by using rapid math operations. By applying the fast coordinate correction method, we can reduce the computational effort within the necessary tolerance of the estimation error. By experiments, the computational effort was improved by up to 99.7% compared to the previous study, and regarding the object’s radar the identification problems were improved by 94.8%. We conclude that the proposed method provides sufficient correcting performance for autonomous driving applications when the multi-sensor coordinates are mismatched.