In order to solve the problem of low safety and efficiency of underground mine vehicles, a path planning method for underground mine vehicles based on an improved A star (A*) and fuzzy control Dynamic Window Approach (DWA) is proposed. Firstly, the environmental information is analyzed, and the proportion of obstacles is introduced into the A* algorithm to optimize the heuristic function to improve its search efficiency in different environments. Secondly, for the global path, the key node selection strategy is proposed, the node selection method is optimized, and the redundant nodes are deleted. The path is optimized by using a 3 times Clamped-B spline to improve the smoothness of the path. Then, based on the principle of fuzzy control, fuzzy rules are established, and a DWA fuzzy controller is designed to improve the environmental adaptability of the DWA algorithm. Finally, on the basis of the global path, the key node is used as the local target point of the DWA, and the fusion of the improved A* and DWA algorithm is realized. Finally, experiments are conducted to verify the effectiveness and feasibility of the proposed path-planning method. The average deviation of the path is controlled at ±0.109 m, which basically meets the path planning needs of underground mine vehicles.

The development of the automobile industry and the increase in car ownership has brought great traffic pressures to the city, among which, the difficulty of parking has become a serious problem to the majority of drivers. An automatic parking system can help drivers to complete parking operation or automatic parking task, and a decision control system is an important part of automatic parking system. In this paper, a strategy for generating the shortest parking path based on a bidirectional breadth-first search algorithm combined with a modified Bellman–Ford algorithm is proposed for automatic parking systems. Experimental results show that this scheme can improve the performance of an automatic parking system, especially in a complex environment.

In order to make up for the shortcomings of independent sensors and provide more reliable estimation, a multi-sensor fusion framework for simultaneous localization and mapping is proposed in this paper. Firstly, the light detection and ranging (LiDAR) point cloud is screened in the front-end processing to eliminate abnormal points and improve the positioning and mapping accuracy. Secondly, for the problem of false detection when the LiDAR is surrounded by repeated structures, the intensity value of the laser point cloud is used as the screening condition to screen out robust visual features with high distance confidence, for the purpose of softening. Then, the initial factor, registration factor, inertial measurement units (IMU) factor and loop factor are inserted into the factor graph. A factor graph optimization algorithm based on a Bayesian tree is used for incremental optimization estimation to realize the data fusion. The algorithm was tested in campus and real road environments. The experimental results show that the proposed algorithm can realize state estimation and map construction with high accuracy and strong robustness.

Medical imaging plays a pivotal role in clinical diagnosis. There are many imaging means to detect the state of tissues and organs in the body. However, these imaging methods have their own advantages and shortcomings. For example, magnetic resonance imaging (MRI) has structural information but no functional characteristics of tissue and positron emission tomography (PET) possesses functional characteristics but no structural information. As we all know, it is helpful for clinical diagnosis to integrate the merits of different modality medical image into a single image. This paper proposed an end-to-end multimodal medical image fusion framework, MMI-fuse. We first applied an auto-encoder to extract the features of source images. Then, an Information preservation-weighted Channel-Spatial attention module (ICS) was proposed to fuse the image features. Finally, we used an auto-decoder modal to restructure the fused medical image. To validate the performance of the proposed method, we collected 106 pairs of brain medical images including MRI-PD and MRI-T2, MRI-PD and SPECT-Tc (technetium-99-sestamibi SPECT), Gadolinium-based MRI and SPECT-Tc, Gadolinium-based MRI and SPECT-Tl (thalium-201 SPECT), as well as MRI and Flurodeoxyglucose PET five sets from Harvard dataset for extensive experiments. In addition, seven methods and five metrics were selected for the comparison experiments. The results demonstrates that the proposed method achieved the best performance on both the visual quality and objective metric score among these seven approaches.

The Rapidly-exploring Random Trees Star (RRT*) is widely used in unmanned vehicle path planning. However, it suffers from excessive sampling randomness, long search time and not capable of dynamic obstacle avoidance. This is not only fatal to safety, but also detrimental to smooth driving. We propose a real-time sampling-based path planning method for unmanned vehicles (RSP-UV) that systematically solves these problems. To address the problem of excessive sampling randomness, we formulate a partitioned sampling strategy and limit the density of sampling points. It reduces the sampling time to a certain extent. Additionally, we design a fuzzy controller to adjust the scale factor of the Dynamic Window Approach (DWA) evaluation function to optimise the local path and improve the dynamic obstacle avoidance efficiency. Fusing the improved RRT* and the improved DWA, the global path output from the improved RRT* is used as the bootstrap path for the improved DWA. Besides, the return path set rule is made to avoid the backtracking phenomenon. This improves the reliability and asymptotic optimality of the final path. Simulation experiments show that the trajectories planned by RSP-UV outperform the three compared methods, the similar method, ORRT*-DWA and MTAR. Real experiments show that the average path length of RSP-UV is reduced by 2.26%, 3.03%, and 3.33% compared to the other three methods, respectively. The average planning time of RSP-UV is reduced by 7.16%, 2.15% and 8.78% compared to the other three methods, respectively. RSP-UV not only ensures the success of path planning but also excels in real-time and adaptability. The paths formed by RSP-UV are cheaper and safer.

With the comprehensive development of intelligent coal mining, research into image enhancement techniques for underground coal mine environments is gaining increasing popularity. This paper proposes a reliable image enhancement method, RIE-CMU, designed for low-light conditions in underground coal mines. It addresses issues such as poor dynamic range expansion, over-enhancement and colour distortion associated with the application of traditional Retinex-Net algorithms in underground coal mines, aiming to systematically resolve these problems. To address the challenge of feature extraction, we have designed a Semantic Embedding Feature Aggregation U-Net (SEFAU-Net) as the decomposition network. By incorporating techniques such as channel attention mechanisms (CA), deformable convolutions, selective kernel feature fusion (SKFF), and PixelShuffle pixel reorganisation, we have significantly enhanced feature extraction capabilities. Furthermore, we have designed a DenseUnet with Convolutional Architecture (DFC-Net) as the image fusion network. We fuse the results of contrast-limited adaptive histogram equalisation (CLAHE) applied to low-light images with the reflection component. This optimises the image enhancement effect and reduces colour bias. By combining the improved decomposition network with the improved fusion network, we utilise the reflection and illumination components output by the decomposition network as inputs to the fusion network. Concurrently, we formulate a multi-task loss function with collaborative constraints to prevent error accumulation. This enhances the clarity and colour fidelity of the final image. The effectiveness of RIE-CMU has been validated through testing on the LOL dataset and our self-built underground coal mine dataset, UCM. Experimental results demonstrate that, in terms of image enhancement quality, RIE-CMU achieves the highest PSNR and SSIM values compared to several benchmark methods, including KinD, Zero-DCE++, RRDNet, R2RNet, URetinex-Net++, and Retinex-Net. The PSNR and SSIM values of RIE-CMU are 22.43% and 70.81% higher, respectively, than those of the unmodified Retinex-Net. The LPIPS and NIQE values of RIE-CMU are reduced by 54.27% and 32.52%, respectively, compared to those of Retinex-Net. Additionally, ablation study and computational complexity analysis further validate the effectiveness of each module and the method's adaptability to underground coal mines.