In this paper, we propose a deep learning based vehicle trajectory prediction technique which can generate the future trajectory sequence of surrounding vehicles in real time. We employ the encoder-decoder architecture which analyzes the pattern underlying in the past trajectory using the long short-term memory (LSTM) based encoder and generates the future trajectory sequence using the LSTM based decoder. This structure produces the K most likely trajectory candidates over occupancy grid map by employing the beam search technique which keeps the K locally best candidates from the decoder output. The experiments conducted on highway traffic scenarios show that the prediction accuracy of the proposed method is significantly higher than the conventional trajectory prediction techniques.

In this paper, we propose an efficient vehicle trajectory prediction framework based on recurrent neural network. Basically, the characteristic of the vehicle's trajectory is different from that of regular moving objects since it is affected by various latent factors including road structure, traffic rules, and driver's intention. Previous state of the art approaches use sophisticated vehicle behavior model describing these factors and derive the complex trajectory prediction algorithm, which requires a system designer to conduct intensive model optimization for practical use. Our approach is data-driven and simple to use in that it learns complex behavior of the vehicles from the massive amount of trajectory data through deep neural network model. The proposed trajectory prediction method employs the recurrent neural network called long short-term memory (LSTM) to analyze the temporal behavior and predict the future coordinate of the surrounding vehicles. The proposed scheme feeds the sequence of vehicles' coordinates obtained from sensor measurements to the LSTM and produces the probabilistic information on the future location of the vehicles over occupancy grid map. The experiments conducted using the data collected from highway driving show that the proposed method can produce reasonably good estimate of future trajectory.

In this paper, we propose a new autonomous braking system based on deep reinforcement learning. The proposed autonomous braking system automatically decides whether to apply the brake at each time step when confronting the risk of collision using the information on the obstacle obtained by the sensors. The problem of designing brake control is formulated as searching for the optimal policy in Markov decision process (MDP) model where the state is given by the relative position of the obstacle and the vehicle's speed, and the action space is defined as the set of the brake actions including 1) no braking, 2) weak, 3) mid, 4) strong brakiong actions. The policy used for brake control is learned through computer simulations using the deep reinforcement learning method called deep Q-network (DQN). In order to derive desirable braking policy, we propose the reward function which balances the damage imposed to the obstacle in case of accident and the reward achieved when the vehicle runs out of risk as soon as possible. DQN is trained for the scenario where a vehicle is encountered with a pedestrian crossing the urban road. Experiments show that the control agent exhibits desirable control behavior and avoids collision without any mistake in various uncertain environments.

In this paper, we deal with comparative evaluation of dynamic and kinematic vehicle models for autonomous driving control systems. Comparative lane coefficients estimation performance of each model is evaluated according to the tire slip angle. Each estimation makes use of clothoidal road constraints, respectively. We observe that the kinematic model is robust against unknown vehicle parameters tire-road condition or cornering stiffness for autonomous driving systems under lane keeping situation. The reliability of each model was validated via not only computational simulation results but also experimental results with a test vehicle.

In this study, we propose a robust nonlinear control method using the barrier Lyapunov function (BLF) under the lateral offset error constraint for lateral control of autonomous vehicles. Furthermore, for the application of the BLF-based control method to lateral control, we propose a second-order lateral dynamics scheme based on the look-ahead distance of the vehicle. In the second-order lateral dynamics, the system functions with unknown parameters, and the external disturbances can be lumped into a disturbance term. The proposed method consists of an extended state observer (ESO) and a nonlinear controller. The ESO is designed to estimate the full state variable and disturbance, including the system modeling and external disturbance. The nonlinear controller is developed using the BLF to compensate for the disturbances and to guarantee the constraint for the lateral offset error. Consequently, the proposed method satisfies the output constraint in the presence of disturbances and modeling uncertainties as well as improves the lateral control performance using only the lateral offset error at the look-ahead distance. The lateral control performance of the proposed method is validated using CarSim and MATLAB/Simulink.

In this paper, a novel multirate lane-keeping system (LKS) is introduced by using a kinematic-based model considering a look-ahead distance. First, we developed a kinematic lateral motion model for LKS with respect to the road. To increase a system damping of vehicle, the unique look-ahead output measurement matrix was introduced. The decentralized multirate lane-keeping control scheme with a multirate Kalman filter was also applied to resolve the asynchronous and irregular sampling time of multi sensors in autonomous vehicles. We validated the performance of the proposed kinematic-based multirate LKS using the dynamic vehicle solver CarSim, MATLAB/Simulink, and electric power steering hardware-in-the-loop system. Experimental validation was also conducted using a real test vehicle. Both simulation and experimental results showed that the proposed kinematic-based LKS performed lane-keeping capability comparable to that of dynamic-based LKS. In particular, the experimental results showed that the proposed method, compared to a human driver, can also maintain the lane within reasonable specifications. The experimental validation was performed at vehicle speeds of up to 120 km/h.