Current lidar-inertial SLAM algorithms mainly rely on the geometric features of the lidar for point cloud alignment. The issue of incorrect feature association arises because the matching process is susceptible to influences such as dynamic objects, occlusion, and environmental changes. To address this issue, we present a lidar-inertial SLAM system based on the LIO-SAM framework, combining semantic and geometric constraints for association optimization and keyframe selection. Specifically, we mitigate the impact of erroneous matching points on pose estimation by comparing the consistency of normal vectors in the surrounding region. Additionally, we incorporate semantic information to establish semantic constraints, further enhancing matching accuracy. Furthermore, we propose an adaptive selection strategy based on semantic differences between frames to improve the reliability of keyframe generation. Experimental results on the KITTI dataset indicate that, compared to other systems, the accuracy of the pose estimation has significantly improved.

Long-term station-keeping of stratospheric airships is challenged by strong time-varying wind fields, pronounced vertical stratification of wind speed and direction, and limited onboard energy. Existing reinforcement-learning approaches typically rely on instantaneous observations to make reactive decisions and therefore struggle to deliver foresighted control in dynamic environments. This paper proposes a Spatio-Temporal Foresight Reinforcement-Learning framework (STF-RL) that explicitly incorporates future wind information. A Transformer is introduced to model multi-step, multi-altitude forecast wind sequences, and a time–height dual positional encoding is designed to characterize both the temporal evolution and the vertical structure of the wind field. A task-conditioned attention pooling mechanism then extracts the future-wind features most relevant to the current state, which are concatenated with the airship state and fed into an actor–critic network to enable foresighted policy learning. A continuous action space supporting three-dimensional maneuvering is constructed, together with a multi-objective reward that jointly accounts for station-keeping performance, energy consumption and safety. Experimental results show that the proposed method outperforms baseline approaches in station-keeping performance, trajectory stability and energy-utilization efficiency, while exhibiting strong robustness across different wind-field conditions.

In recent years, with the continuous development of the global economy, maritime transportation has become an essential mode of transportation for both domestic and international trade, making it an urgent task for countries around the world to improve the intelligence level of maritime traffic management. As a key technology in real-time vessel monitoring, traffic hazard early warning, and traffic flow estimation, vessel trajectory prediction has been widely applied across both civil and commercial sectors. To address the problems with existing trajectory prediction methods, such as difficulty in establishing real-time vessel kinematic models, limited understanding of vessel navigation strategies, and poor adaptability for long-term prediction, this paper proposes a deep reinforcement learning-based vessel trajectory prediction method. Firstly, the trajectory prediction problem is formulated as a Markov Decision Process (MDP), thereby transforming the prediction problem into an optimal policy-solving problem of the MDP. Secondly, given the complexity of the trajectory prediction problem, a convolutional neural network (CNN) is adopted to parameterize the policy network, and a CNN-PPO deep reinforcement learning method is used to learn the target vessel’s navigation strategy from historical trajectories. Comparative experimental results show that the proposed algorithm is not only suitable for predicting the target’s position at a specific future moment but also offers clear advantages in trajectory prediction across multiple future moments.