Reinforcement learning has gradually demonstrated its decision-making ability in autonomous driving. Reinforcement learning is learning how to map states to actions by interacting with environment so as to maximize the long-term reward. Within limited interactions, the learner will get a suitable driving policy according to the designed reward function. However there will be a lot of unsafe behaviors during training in traditional reinforcement learning. This paper proposes a RL-based method combined with RL agent and Monte Carlo tree search algorithm to reduce unsafe behaviors. The proposed safe reinforcement learning framework mainly consists of two modules: risk state estimation module and safe policy search module. Once the future state will be risky calculated by the risk state estimation module using current state information and the action outputted by the RL agent, the MCTS based safe policy search module will activate to guarantee a safer exploration by adding an additional reward for risk actions. We test the approach in several random overtake scenarios, resulting in faster convergence and safer behaviors compared to traditional reinforcement learning.

We propose a novel, computationally efficient, and robust light detection and ranging (LiDAR)-only odometry framework based on truncated least squares termed T-LOAM. Our method focuses on alleviating the impact of outliers to allow robust navigation in sparse, noisy, or cluttered scenarios where degeneration occurs. As preprocessing, the multiregion ground extraction and dynamic curved-voxel clustering methods are proposed to accomplish the segmentation of 3D point clouds and filter out unstable objects. A novel feature extraction module is tailored to discriminate four peculiar features: edge features, sphere features, planar features, and ground features. As frontend, a hierarchical feature-based LiDAR-only odometry performs precise motion estimates through the truncated least squares method for directly processing various features. The preprocessing model and motion estimation precision have been evaluated on the KITTI odometry benchmark as well as various campus scenarios. The experimental results have demonstrated the real-time capability and superior precision of the proposed T-LOAM over other state-of-the-art algorithms.

As an indispensable branch of machine learning (ML), reinforcement learning (RL) plays a prominent role in the decision-making process of autonomous driving (AD), which enables autonomous vehicles (AVs) to learn an optimal driving strategy through continuous interaction with the environment. This paper proposes a deep reinforcement learning (DRL)-based motion planning strategy for AD tasks in the highway scenarios where an AV merges into two-lane road traffic flow and realizes the lane changing (LC) maneuvers. We integrate the DRL model into the AD system relying on the end-to-end learning method. An improved DRL algorithm based on deep deterministic policy gradient (DDPG) is developed with well-defined reward functions. In particular, safety rules (SR), safety prediction (SP) module and trauma memory (TM) as well as the dynamic potential-based reward shaping (DPBRS) function are adopted to further enhance safety and accelerate learning of the LC behavior. For validation, the proposed DSSTD algorithm is trained and tested on the dual-computer co-simulation platform. The comparative experimental results show that our proposal outperforms other benchmark algorithms in both driving safety and efficiency.

Path tracking as an essential function of an autonomous vehicle ensures it tracks the desired path. Its performance could be degraded by wind, road conditions, vehicle parameter uncertainty and signal delay. This study proposes a path tracking control framework for autonomous vehicle by combining tube model predictive control (MPC) and time delay motion prediction (TDMP). Tube MPC with N ‐step reachable set is proposed to handle disturbances. A robust invariant set is computed efficiently using supporting function and guarantees state constraints in the presence of disturbances. In TDMP, a path during the signal delay is presented and a vehicle kinematics model is adopted to predict changes of vehicle position and yaw. Simulations show that the computational cost with support function for a robust positively invariant set is less than that with Multi‐Parametric Toolbox. In addition, the proposed control framework not only addresses disturbances like wind, road conditions and vehicle parameter uncertainty, but also handles the steering signal delay through motion prediction.

Traditional decoupling trajectory planning strategy deals with complex urban road scenes with many shortcomings, such as stereotyped behavior and poor trajectory quality. Aiming at these problems, the paper proposes a combined spatiotemporal trajectory planning and control method based on reachable set and optimization technique. Firstly, to address the shortcomings of existing reachable set methods in terms of unreasonable risk characterization and lack of risk consideration in the spatiotemporal corridor generation process, a risk assessment model that takes into account the uncertainty of the predicted location distribution and a reachable set spatiotemporal corridor generation strategy that integrates the risk field are proposed. Secondly, a trajectory optimization strategy of rolling iterative optimization within the spatiotemporal corridor is designed by using the iterative linear quadratic regulator (ILQR) and taking into account the dynamics of the vehicle's transverse-longitudinal coupling. The optimization strategy ensures the dynamics of the trajectory is feasible and realizes the strong coupling of planning and control. Finally, a safety check module is designed to coordinate the time-consuming corridor generation and the real-time application of the algorithm. To verify the performance of the proposed method, the long-time simulation and real-vehicle comparison tests of complex stochastic interaction scenes are fully carried out. The test results show that the proposed method can effectively realize the coordination between driving efficiency, comfort, and safety in high-density complex traffic flow.