Simultaneous localization and mapping (SLAM) plays an important role in the state estimation of mobile robots. Most popular LiDAR SLAM (L-SLAM) methods extract feature points only from the geometric structure of the environment, which can result in inaccurate localization in degenerated scenarios. In this article, we present a novel framework for LiDAR intensity gradient enhanced tightly coupled LiDAR-inertial odometry (IGE-LIO). The framework proposes a novel LiDAR intensity gradient-based feature extraction approach for accurate pose estimation, overcoming the challenges faced by L-SLAM in degenerated environments. After computing the intensity gradient of each LiDAR point, we dynamically extract intensity edge points (IEPs) from texture information. In addition, we extract geometric planar points (GPPs) and geometric edge points (GEPs) based on geometric information. Then, the error analysis is performed on each type of feature points, and the weighting functions are designed to correct measurement noise and mitigate biases introduced by the additional uncertainty in feature extraction. Subsequently, an iterative extended Kalman filter (IEKF) framework is constructed by combining residuals from point-to-plane and point-to-edge associations. Finally, extensive experiments are conducted in indoor, outdoor, and LiDAR degenerated scenarios. The results demonstrate the significantly improved robustness and accuracy of our proposed method compared with the existing geometric-only methods, especially in LiDAR degenerated scenarios.

Currently, high-fidelity simulation techniques, such as discrete element methods, are widely employed for predicting the off-road mobility of unmanned ground vehicles, ensuring the successful completion of field missions. However, the traversal simulation process based on discrete element terrain models is highly time-consuming. To address this challenge, we propose a double neural networks enhanced global mobility prediction model. Firstly, acceleration of the vehicle-terrain contact simulation is achieved by developing an artificial neural network surrogate model. This model predicts stress and strain relationships based on a multiscale representative volume element terrain model, effectively replacing complex mechanical calculations. Secondly, the generation of the global mobility map is expedited by introducing a boundary data enhanced artificial neural network model. This model generates a training dataset by simulating Gaussian sampled terrain data and additional enhanced terrain data at the mobility classification boundary. The trained artificial neural network is subsequently utilized to predict the mobility map, thus circumventing the time-consuming process of traversing simulations. The simulation tests validate that our proposed method not only significantly reduces computation time but also ensures high prediction accuracy.

With the rapid advancement of technologies related to unmanned ground systems, ground vehicles are being widely deployed across various domains. However, when operating in complex, soft terrain environments, the low bearing capacity of such terrains poses a significant challenge to vehicle mobility. This paper presents a comprehensive review of mobility prediction methods for ground vehicles in off-road environments. We begin by discussing the concept of vehicle mobility, followed by a systematic and thorough summary of the primary prediction methods, including empirical, semi-empirical, numerical simulation, and machine learning approaches. The strengths and weaknesses of these methods are compared and analyzed in detail. Subsequently, we explore the application scenarios of mobility prediction in military operations, subsea work, planetary exploration, and agricultural activities. Finally, we address several existing challenges in current mobility prediction methods and propose exploratory research directions focusing on key technologies and applications, such as real-time mobility prediction, terrain perception, path planning on deformable terrain, and autonomous mobility prediction for unmanned systems. These insights aim to provide valuable reference points for the future development of vehicle mobility prediction methods.

Three-dimensional (3D) object detection is crucial for autonomous driving, yet current PointPillar feature-based methods face challenges like under-segmentation, overlapping, and false detection, particularly in occluded scenarios. This paper presents a novel dual-stage improved PointPillar feature-based 3D object detection method (S2*-ODM) specifically designed to address these issues. The first innovation is the introduction of a dual-stage pillar feature encoding (S2-PFE) module, which effectively integrates both inter-pillar and intra-pillar relational features. This enhancement significantly improves the recognition of local structures and global distributions, enabling better differentiation of objects in occluded or overlapping environments. As a result, it reduces problems such as under-segmentation and false positives. The second key improvement is the incorporation of an attention mechanism within the backbone network, which refines feature extraction by emphasizing critical features in pseudo-images and suppressing irrelevant ones. This mechanism strengthens the network’s ability to focus on essential object details. Experimental results on the KITTI dataset show that the proposed method outperforms the baseline, achieving notable improvements in detection accuracy, with average precision for 3D detection of cars, pedestrians, and cyclists increasing by 1.04%, 2.17%, and 3.72%, respectively. These innovations make S2*-ODM a significant advancement in enhancing the accuracy and reliability of 3D object detection for autonomous driving.

Bird's eye view (BEV) segmentation map is a recent development in autonomous driving that provides effective environmental information, such as drivable areas and lane dividers. Most of the existing methods use cameras and LiDAR as inputs for segmentation and the fusion of different modalities is accomplished through either concatenation or addition operations, which fails to exploit fully the correlation and complementarity between modalities. This paper presents the CMGFA (Cross-Modal Group-mix attention Feature Aggregator), an end-to-end learning framework that can adapt to multiple modal feature combinations for BEV segmentation. The CMGFA comprises the following components: (i) The camera has a dual-branch structure that strengthens the linkage between local and global features. (ii) Multi-head deformable cross-attention is applied as cross-modal feature aggregators to aggregate camera, LiDAR, and Radar feature maps in BEV for implicit fusion. (iii) The Group-Mix attention is used to enrich the attention map feature space and enhance the ability to segment between different categories. We evaluate our proposed method on the nuScenes and Argoverse2 datasets, where the CMGFA significantly outperforms the baseline. The code will be updated at https://github.com/kuangxk2016/CMGFA/tree/master .