This paper presents a surround vehicle motion prediction algorithm for multi-lane turn intersections using a Long Short-Term Memory (LSTM)-based Recurrent Neural Network (RNN). The motion predictor is trained using the states of subject and surrounding vehicles, which are collected by sensors mounted on an autonomous vehicle. Data on 484 vehicle trajectories were collected from real traffic situations at multi-lane turn intersections. 11,662 and 4,998 samples acquired from the vehicle trajectories were used to train and evaluate the networks, respectively. A motion planner based on Model Predictive Control (MPC) is designed to determine the longitudinal acceleration command based on the predicted states of surrounding vehicles. The future states of the subject vehicle derived by MPC is used as an input feature to reflect the interaction of subject and target vehicles in LSTM-RNN based motion predictor. The proposed algorithm was evaluated in terms of its accuracy and its effects on the motion planning algorithm based on the driving data sets. The improved prediction accuracy substantially increased safety by bounding the prediction error within the safety margin. The application results of the proposed predictor demonstrate the improved recognition timing of the preceding vehicle and the similarity of longitudinal acceleration with drivers.

This paper presents an MPC-based integrated control algorithm for an autonomous vehicle equipped with four-wheel independent steering and driving systems. The objective of this research is to improve the performance of the path and velocity tracking controllers by distributing the control effort to the multiple actuators. The proposed algorithm has two modules: reference state decision and MPC-based vehicle motion controller. Reference state decision module determines reference state profiles consisting of yaw rate and velocity in order to overcome the limitation of the error dynamics-based path tracking controller, which requires several assumptions on the reference path. The MPC-based vehicle motion controller is designed with a linear time-varying vehicle model in order to optimally allocate the control effort to each actuator. A linear time-varying MPC is adopted to reduce computational burden caused by using a non-linear one. The effectiveness of the proposed algorithm is validated via simulation on MATLAB/Simulink and CarSim. The simulation results show that the proposed algorithm improves the reference tracking performance by effectively distributing the control effort to the steering angle and driving force of each actuator.

This paper presents a method to design a path tracking controller with four-wheel independent braking (4WIB), drive (4WID) and steering (4WIS) systems equipped in in-wheel motor-driven electric vehicles (IWM-EVs). Generally, it is difficult to calculate the steering angles of 4WIS and the braking/traction torques of 4WIB/4WID for path tracking control. Moreover, there have been limitations of an error dynamics-based path tracking controller, which requires assumptions on a target path. To cope with these problems, the path tracking problem on a target path is converted into the yaw rate tracking one with a reference yaw rate in this paper. Two methods are adopted for the purpose of calculating a reference yaw rate. The first is to use a pure pursuit method, which generates a steering angle for path tracking. From the steering angle, a reference yaw rate is calculated. The second is to derive a reference yaw rate from a target path and a current vehicle position. For yaw rate tracking, direct yaw moment control is adopted to generate a control yaw moment. A control allocation method is adopted to distribute a control yaw moment into tire forces, generated by 4WIS, 4WID and 4WIB. Several actuator combinations are represented by various sets of virtual weights in the control allocation method. A simulation with a vehicle simulation program, CarSim ^® , shows that the proposed path tracking controller is effective in enhancing the path tracking performance with 4WIS, 4WID and 4WIB. From the simulation, effects of actuator combinations on path tracking performance are analyzed.

Detection and tracking of moving objects (DATMO) in an urban environment using Light Detection and Ranging (LiDAR) is a major challenge for autonomous vehicles due to sparse point cloud, multiple moving directions, various traffic participants, and computational load. To address the complexity of this issue, this study presents a novel model-free approach for DATMO using 2D LiDAR implemented on autonomous vehicles. The approach has been used to classify moving points in the point cloud using the predicted Static Obstacle Map (SOM) generated via interaction between Geometric Model-Free Approach (GMFA) and SOM, and estimates the state of each moving object via GMFA. The motion of each point represented by the state of moving objects updates the SOM. The interaction between GMFA and SOM estimates the correspondence between consecutive point clouds in real-time. The proposed approach has been evaluated via RT range and labeled dataset. The accuracy of estimation of the yaw angle and the velocity of a moving vehicle has been quantitatively evaluated using the RT-range. The performance is significantly improved compared with the geometric model-based tracking (MBT). The estimation of the yaw angle, which has a significant effect on the cut-in/cut-out intention of the target vehicle, is shown to be remarkably improved. Based on the evaluation of the labeled dataset, false-positive and false-negative features are suppressed more than MBT.

This paper presents a method by which to design linear quadratic (LQ) static output feedback (SOF) controllers with a 2-DOF quarter-car model for an active suspension system. Generally, it is challenging to implement linear quadratic regulator (LQR), designed with a full-car model, in actual vehicles because doing so requires 14 state variables to be precisely measured. For this reason, LQR has been designed with a quarter-car model and then applied to a full-car model. Although this requires far fewer state variables, some of them are still difficult to measure. Thus, it is necessary to design a LQ SOF controller which uses available sensor signals that are relatively easily measured in real vehicles. In this paper, a LQ SOF controller is designed with a quarter-car model and applied to a full-car model for ride comfort. To design the controller, an optimization problem is formulated and solved by a heuristic optimization method. A frequency domain analysis and a simulation with a simulation package show that the proposed LQ SOF controllers effectively improve the ride comfort with an active suspension system.