The paper presents a vision-based obstacle avoidance strategy for lightweight self-driving cars that can be run on a CPU-only device using a single RGB-D camera. The method consists of two steps: visual perception and path planning. The visual perception part uses ORBSLAM3 enhanced with optical flow to estimate the car’s poses and extract rich texture information from the scene. In the path planning phase, the proposed method employs a method combining a control Lyapunov function and control barrier function in the form of a quadratic program (CLF-CBF-QP) together with an obstacle shape reconstruction process (SRP) to plan safe and stable trajectories. To validate the performance and robustness of the proposed method, simulation experiments were conducted with a car in various complex indoor environments using the Gazebo simulation environment. The proposed method can effectively avoid obstacles in the scenes. The proposed algorithm outperforms benchmark algorithms in achieving more stable and shorter trajectories across multiple simulated scenes.

Safety and efficiency are crucial for autonomous driving in roundabouts, especially mixed traffic with both autonomous vehicles (AVs) and human-driven vehicles. This paper presents a learning-based algorithm that promotes safe and efficient driving across varying roundabout traffic conditions. A deep Q-learning network is used to learn optimal strategies in complex multi-vehicle roundabout scenarios, while a Kolmogorov-Arnold Network (KAN) improves the AVs’ environmental understanding. To further enhance safety, an action inspector filters unsafe actions, and a route planner optimizes driving efficiency. Moreover, model predictive control ensures stability and precision in execution. Experimental results demonstrate that the proposed system consistently outperforms state-of-the-art methods, achieving fewer collisions, reduced travel time, and stable training with smooth reward convergence.

Semantic segmentation plays a critical role in enabling intelligent vehicles to comprehend their surrounding environments. However, deep learning-based methods usually perform poorly in domain shift scenarios due to the lack of labeled data for training. Unsupervised domain adaptation (UDA) techniques have emerged to bridge the gap across different driving scenes and enhance model performance on unlabeled target environments. Although self-training UDA methods have achieved state-of-the-art (SOTA) results, the challenge of generating precise pseudo-labels persists. These pseudo-labels tend to favor majority classes, consequently sacrificing the performance of rare classes or small objects like traffic lights and signs. To address this challenge, we introduce SAM4UDASS, a novel approach that incorporates the Segment Anything Model (SAM) into self-training UDA methods for refining pseudo-labels. It involves Semantic-Guided Mask Labeling, which assigns semantic labels to unlabeled SAM masks using UDA pseudo-labels. Furthermore, we devise fusion strategy aimed at mitigating semantic granularity inconsistency between SAM masks and the target domain. SAM4UDASS innovatively integrate segmentation foundation model SAM with UDA for semantic segmentation in driving scenes and seamlessly complements self-training UDA methods. Extensive experiments on synthetic-to-real, normal-to-adverse, and pinhole-to-panoramic driving datasets demonstrate its effectiveness. It brings more than 3% mIoU gains on GTA5-to-Cityscapes, SYNTHIA-to-Cityscapes, and Cityscapes-to-ACDC when using DAFormer and achieves SOTA when using MIC. The code is available at https://github.com/ywher/SAM4UDASS .

Experiments under two intake air swirl levels (swirl ratios of 0.55 and 5.68) were conducted in order to investigate the early flame development of combustion in a single-cylinder spark-ignition direct-injection engine. The engine was equipped with a quartz insert in the piston, which provided an optical access to its cylinder through the piston. The crank angle resolved combustion images through the piston window and in-cylinder pressure measurements of 250 cycles were simultaneously recorded for both swirl levels at a specified engine speed and low load condition. The early development, size, and spatial characteristics extracted from the flame images were analyzed as a function of crank angle degrees after the ignition. The experimental results revealed that the early flame development was strongly influenced by the highly directed swirl motion of intake-air into the combustion cylinder. The location of the start of the flame kernel relative to the spark plug position also changed intermittently at different swirl levels. While the structure of the early flame was found to be similar for both swirl levels, the starting location of the flame showed a vast difference in how the flame progressed. In general, the flame kernel was formed two crank-angle degrees after spark timing for the high swirl level, which was four crank-angle degrees earlier than that of the low swirl case. For the low swirl flow, the early combustion showed more cycle-to-cycle variation in terms of both the flame size and centroid location. It was quantitatively shown that increasing the swirl ratio from 0.55 to 5.68 could reduce the cycle-to-cycle variation of the early flame structure, resulting in about three to four crank-angle degrees advance of the peak pressure location and a 1% improvement for the coefficient of variation (COV) of the indicated mean effective pressure (IMEP).

Feature descriptors, as abstraction of the critical information in the lidar point clouds, are often used in global pose initialization in large-area to provide a pose reference for intelligent driving system. The current state-of-the-art method Scan Context descriptor is generated based on points’ maximum height and, therefore, designed especially for outdoor scenarios without a ceiling. This study proposes a generic descriptor for both outdoor and indoor scenarios based on the point cloud’s Gridded Gaussian Distribution (GGD). Wasserstein distance is introduced to this field to evaluate the proposed GGD descriptors’ matching performance because it not only has a solid mathematical foundation in comparing two Gaussian distributions but also shows excellent time efficiency via a straightforward analytical solution without enumeration or iteration. We construct a multi-step error function to initialize the vehicle pose using conventional cosine similarity, and Wasserstein distance. Two experiments are designed for this study. The first experiment compares the pose initialization performances under various multi-frame superimposition distance in space to find an efficient GGD descriptor extraction setting. The second experiment verifies that the proposed method achieves a better pose initialization success rate than the mainstream methods.