This paper describes an integrated autonomous driving (AD) control system for an autonomous vehicle with four independent in-wheel motors (IWMs). The system consists of two parts: the AD controller and the chassis controller. These elements are functionally integrated to improve vehicle stability and path tracking performance. The vehicle is assumed to employ an IWM independently at each wheel. The AD controller implements longitudinal/lateral path tracking using proportional-integral(PI) control and adaptive model predictive control. The chassis controller is composed of two lateral control units: the active front steering (AFS) control and the torque vectoring (TV) control. Jointly, they find the yaw moment to maintain vehicle stability using sliding mode control; AFS is prioritized over TV to enhance safety margin and energy saving. Then, the command yaw moment is optimally distributed to each wheel by solving a constrained least-squares problem. Validation was performed using simulation in a double lane change scenario. The simulation results show that the integrated AD control system of this paper significantly improves the path tracking capability and vehicle stability in comparison with other control systems.

In truck platooning, the leading vehicle is driven manually, and the following vehicles run by autonomous driving, with the short inter-vehicle distance between trucks. To successfully perform platooning in various situations, each truck must maintain dynamic stability, and furthermore, the whole system must maintain string stability. Due to the short front-view range, however, the following vehicles’ path planning capabilities become significantly impaired. In addition, in platooning with articulated cargo trucks, the off-tracking phenomenon occurring on a curved road makes it hard for the following vehicle to track the trajectory of the preceding truck. In addition, without knowledge of the global coordinate system, it is difficult to correlate the local coordinate systems that each truck relies on for sensing environment and dynamic signals. In this paper, in order to solve these problems, a path planning algorithm for platooning of articulated cargo trucks has been developed. Using the Kalman filter, V2V (Vehicle-to-Vehicle) communication, and a novel update-and-conversion method, each following vehicle can accurately compute the trajectory of the leading vehicle’s front part for using it as a target path. The path planning algorithm of this paper was validated by simulations on severe driving scenarios and by tests on an actual road. The results demonstrated that the algorithm could provide lateral string stability and robustness for truck platooning.

Deep learning algorithms are widely adopted in autonomous driving due to their strong capabilities of classifying the objects around the vehicle. This paper introduces methodologies for constructing datasets for training deep learning algorithms that receive LiDAR sensor data. The training datasets were built in a virtual environment using 3D LiDAR sensor models and different object models. These datasets were used to train the simple CNN model developed in this study. The performance of the trained CNN model was evaluated using the Waymo open datasets and driving scenarios that include multiple moving objects interfering with one another. The CNN model proved to be as good as the best benchmark models, which in turn assured the validity of the training datasets in this study. The proposed method can achieve significant time and cost savings in the generation of proper training datasets for deep learning algorithms in a variety of autonomous driving complexities.