Search

In the field of autonomous driving, the detection of targets such as vehicles, bicycles, and pedestrians in complex road conditions is of great importance. Through extensive experimentation, we have found that various vehicle targets generally occupy large sizes in the image but are easily occluded, while small targets such as pedestrians usually appear densely. The detection of targets of different sizes is an important challenge for the performance of current detectors. To address this issue, we proposed a novel hierarchical feature pyramid network structure. This structure comprises a series of CNN-Transformer variant layers, each of which is a superposition of CST neural network modules and Swin Transformer modules. In addition, considering that the huge computation of the global self-attention mechanism is difficult to be applied in the field of autonomous driving, we adopted the shifted window method in SwinFM, which effectively accelerates the inference process by replacing the traditional method by using the self-attention mechanism within the window. This study uses the Swin Transformer as a baseline. Compared to the baseline, our EDSD model improves the average accuracy by 1.8% and 3.1% on the BDD100K dataset and the KITTI dataset, respectively. [ABSTRACT FROM AUTHOR]

AbstractThis study presents the outcomes of empirical investigations pertaining to human-vehicle interactions involving an autonomous vehicle (AV) equipped with both internal and external Human Machine Interfaces (HMIs) within a crosswalk scenario. The internal and external HMIs were integrated with implicit communication techniques, incorporating a combination of gentle and aggressive braking manoeuvres within the crosswalk. Data were collected through a combination of questionnaires and quantifiable metrics, including pedestrian decision to cross related to the vehicle distance and speed. The questionnaire responses reveal that pedestrians experience enhanced safety perceptions when the external HMI and gentle braking manoeuvres are used in tandem. In contrast, the measured variables demonstrate that the external HMI proves effective when complemented by the gentle braking manoeuvre. Furthermore, the questionnaire results highlight that the internal HMI enhances passenger confidence only when paired with the aggressive braking manoeuvre. [ABSTRACT FROM AUTHOR]

Introduction: Intelligent vehicles and autonomous driving have been the focus of research in the field of transport, but current autonomous driving models have significant errors in lateral tracking that cannot be ignored. Methods: In view of this, this study innovatively proposes a lateral trajectory algorithm for intelligent vehicles based on improved radial basis function (RBF). The algorithm first models the lateral trajectory behaviour of the car based on the pre-scanning steering theory, and then proposes an improved RBF network model to compensate for the error of the lateral trajectory model and further improve the accuracy. Results: According to the simulation test results, after 20 iterations, the proposed algorithm always shows the highest accuracy with the same number of iterations. When the number of iterations reaches 370, the accuracy of the algorithm is stable at 88%. In addition, the bending test shows that the proposed algorithm performs best at low speeds with an overall error of 0.028 m, which is a higher accuracy compared to the algorithm without neural network compensation. Discussion: The maximum error of the proposed algorithm does not exceed 0.04 m in complex continuous curved terrain, which is safe within the normal road width. Overall, the lateral tracking algorithm proposed in this research has better lateral tracking capability compared to other improved algorithms of the same type. The research results are of some significance to the field of lateral tracking of automatic driving, which provides new ideas and methods for the field of lateral tracking of automatic driving technology and helps to promote the overall development of automatic driving technology. By reducing the lateral tracking error, the driving stability and safety of the self-driving car can be improved, creating favourable conditions for the wide application of the self-driving technology. [ABSTRACT FROM AUTHOR]

Recently, the intelligent in-vehicle voice agent (IVVA) using natural language processing (NLP) and capable of having a conversation have been introduced in autonomous vehicles (AVs). The IVVA is expanding its role not only to provide driving information and vehicle condition updates to ensure safety, but also to communicate and empathize with drivers like a friend in order to provide a more enjoyable driving experience. Accordingly, various anthropomorphic techniques have been applied to IVVAs and their effects evaluated. There is a tendency to focus on identifying whether anthropomorphic techniques are effective or not, but consideration of the autonomous driving contexts (ADCs) has been insufficient. Therefore, this study compares and evaluates the effects of the ADC (e.g., emergency stop, navigation, and casual conversations) on the interaction experience (specifically, intimacy, trust, and intention to use) from human-IVVA conversations by analyzing two IVVAs with different levels of anthropomorphism. As a result, the IVVA with the higher level of anthropomorphism encouraged greater intimacy. An interaction effect was confirmed based on the ADC and the level of anthropomorphism. In addition, regarding trust and intention to use, the IVVA with greater anthropomorphism was evaluated with a higher trust level and a stronger intention to use in an emergency stop situation. But the IVVAs did not always provide a positive experience in other driving contexts. The results of this study suggest that when designing an IVVA for AVs, it is necessary to use a conversation strategy appropriate to the situation by recognizing the ADC, rather than simply considering an increase in anthropomorphism. [ABSTRACT FROM AUTHOR]

As a new, highly complex, and far‐reaching technology, autonomous driving can be associated with various fears and uncertainties. However, recent findings show that high trait anxiety can positively contribute to the intention to use (ITU) autonomous vehicles (AVs). An explanation for this is that the possibility of handing over one's driving control to artificial intelligence (AI) is even more relieving for more anxious people. Our study aimed to test whether this explanation can be supported by investigating to what extent this relationship can be applied to buses in which control is handed over per se–in the conventional bus to a driver, and in the autonomous bus to the AI. We also analyzed how the fear of giving up control mediates the relationship between trait anxiety and ITU. In a quasi‐experimental study, 253 subjects were surveyed while riding an autonomous or conventional electric bus. The results confirmed a positive association between trait anxiety and ITU in the overall sample but not in the autonomous and conventional subsamples. Contrary to our assumptions, fear of giving up control served as a slightly suppressive but not significant mediator. The results were independent of whether control was handed over to a human driver in the conventional electric bus or to AI in the autonomous bus. Our study thus provides fundamental new insights into the acceptance of AVs and buses in general and opens the door for subsequent research based on these findings. [ABSTRACT FROM AUTHOR]

Predicting the mobility patterns of vehicles together with their interactions among surrounding traffic objects is a critical task in autonomous driving systems. Existing graph neural network-based trajectory prediction models primarily capture the structural connectivity of network nodes (road objects) and assume equal priority to all neighbors of a node. However, in real-time traffic networks, the behavior of each vehicle is significantly influenced by its neighboring road objects and this influence is not uniform. This necessitates a neighbor interaction-aware trajectory prediction model that assumes non-uniform priority among neighboring nodes. In this article, we have designed a novel unified trajectory prediction model which is suitable for both highway and urban traffic conditions. The proposed approach seamlessly integrates multi-level context modeling using graph attention mechanisms, capturing and leveraging interactions and dependencies among objects at varied levels of proximity within a graph. Additionally, it employs an encoder–decoder long short-term memory architecture for long-term trajectory prediction, ensuring adaptability to different driving scenarios. The advanced graph attention mechanisms play a crucial role in modeling spatial dependencies between vehicles, allowing the proposed model to dynamically adapt to evolving interactions over time. The experimentations done on real-world trajectory datasets, namely, Next Generation Simulation US-101 highway dataset and diverse urban datasets such as ApolloScape and Argoverse demonstrate remarkable performance of MC-GATP in long-term trajectory prediction. The model showcases superior prediction accuracy, scalability, and computational efficiency for both highway and urban environments. [ABSTRACT FROM AUTHOR]

The continuous advancement of autonomous driving technology imposes higher demands on the accuracy of target detection in complex environments, particularly when traffic targets are occluded. Existing algorithms still face significant challenges in detection accuracy and real-time performance under such conditions. To address this issue, this paper proposes an improved YOLOX algorithm based on adaptive deformable convolution, named OCC-YOLOX. This algorithm enhances the feature extraction network's ability to focus on occluded targets by incorporating a coordinate attention mechanism. Additionally, it introduces the Overlapping IoU (OL-IoU) loss function to optimize the overlap between predicted and ground truth bounding boxes, thereby improving detection accuracy. Furthermore, the adoption of Fast Spatial Pyramid Pooling (Fast SPP) reduces computational complexity while maintaining real-time performance. Experiments on fused public datasets demonstrate that OCC-YOLOX achieves improvements in accuracy, recall, and average precision by 2.76%, 1.25%, and 1.92%, respectively. In addition to testing on the KITTI, CityPersons, and BDD100K datasets, the effectiveness of the OCC-YOLOX algorithm is further validated through comparisons with self-collected occlusion scene data. The experimental results indicate that OCC-YOLOX outperforms existing mainstream detection algorithms, particularly in handling complex occlusion scenarios, significantly enhancing the accuracy and efficiency of object detection. This study provides new insights for addressing the challenges of occluded target detection in intelligent transportation systems. [ABSTRACT FROM AUTHOR]

ABSTRACT In this study, we developed a four‐wheel‐driven autonomous mobile robot to transport heavy objects on narrow alleys and steep slopes. The robot, which weighs 450 kg, is driven by four 650 W hub motors to climb 20° steep slopes and travel through a narrow alley. Two 2D‐LiDARs attached to the front and rear of the robot body are used for highly accurate driving. A method based on the least squares approach for estimating the position and posture of the robot body is proposed by using the point cloud information of the side wall of the narrow alley. Field experiments on test courses demonstrated that the autonomous robot could transport heavy objects over steep slopes and travel in a narrow alley with high accuracy. [ABSTRACT FROM AUTHOR]

3D point clouds collected by low-channel light detection and ranging (LiDAR) are relatively sparse compared to high-channel LiDAR, which is considered costly. To address this, an outdoor large-scene point cloud reconstruction (LSPCR) technique based on transformer is proposed in this study. The LSPCR approach first projects the original sparse 3D point cloud onto a 2D range image; then, it enhances the resolution in the vertical direction of the 2D range image before converting the high-resolution range image back to a 3D point cloud as the final reconstructed point cloud data. Experiments were performed on the real-world KITTI dataset, and the results show that LSPCR achieves an average accuracy improvement of over 60% compared to non-deep-learning algorithms; it also achieves better performance compared to the latest deep-learning algorithms. Therefore, LSPCR is an effective solution for sparse point cloud reconstruction and addresses the challenges associated with high-resolution LiDAR point clouds. [ABSTRACT FROM AUTHOR]

Addressing the issue of inadequate dynamic object detection accuracy in current road driving warning systems, this study proposes the RepBF‐YOLOv8 detection algorithm aimed at efficient risk identification. The backbone network of YOLOv8n is replaced with the lightweight RepViT architecture, which is more suitable for visual tasks. This replacement simplifies the traditional structure, reduces the complexity of the backbone network, maximizes performance enhancement, and minimizes latency. Additionally, the FPN in the neck section is upgraded to Bi‐FPN, which reduces nodes and span connections and incorporates rapid normalization to achieve fast multi‐scale feature fusion. For risk grading, the algorithm infers distances and collision times, categorizing detected objects into high, medium, and low‐risk levels, and uses different colors to warn the driver. Comparative experimental results show that the optimized algorithm improves Precision by 1.7%, Recall by 2.3%, mAP@0.5 by 1.53%, and mAP@0.5:0.95 by 2.91%. In road tests, the risk warning system achieves a frame detection rate ranging from a minimum of 38.4 fps to a maximum of 59.0 fps. The detection confidence for various objects remains above 0.71, reaching as high as 0.98. Specifically, the "Car" confidence ranges from 0.81 to 0.98, demonstrating the accuracy and robustness of vehicle risk detection. [ABSTRACT FROM AUTHOR]

An accurate panorama depth estimation result is crucial to risk perception in autonomous driving practice. In this paper, an innovative framework is presented to address the challenges of imperfect observation and projection fusion in panorama depth estimation, enabling the accurate capture of distances from surrounding images in driving scenarios. First, the Patch Filling method is proposed to alleviate the imperfect observation of panoramic depth in autonomous driving scenarios, which constructs a panoramic depth map based on the sparse distance data provided by the 3D point cloud. Then, in order to tackle the distortion challenge faced by outdoor panoramic images, a method for image context learning, ViT-Fuse, is proposed and specifically designed for equirectangular panoramic views. The experimental results show that the proposed ViT-Fuse reduces the estimation error by 9.15% on average in driving scenarios compared with the basic method and exhibits more robust and smoother results on the edge details of the depth estimation maps. [ABSTRACT FROM AUTHOR]

The current 3D object detection methods have achieved promising results for conventional tasks to detect frequently occurring objects like cars, pedestrians and cyclists. However, they require many annotated boundary boxes and class labels for training, which is very expensive and hard to obtain. Nevertheless, detecting infrequent occurring objects, such as police vehicles, is also essential for autonomous driving to be successful. Therefore, we explore the potential of few-shot learning to handle this challenge of detecting infrequent categories. The current 3D object detectors do not have the necessary architecture to support this type of learning. Thus, this paper presents a new method termed few-shot single-stage network for 3D object detection (FS-3DSSN) to predict infrequent categories of objects. FS-3DSSN uses a class-incremental few-shot learning approach to detect infrequent categories without compromising the detection accuracy of frequent categories. It consists of two modules: (i) a single-stage network architecture for 3D object detection (3DSSN) using deformable convolutions to detect small objects and (ii) a class-incremental-based meta-learning module to learn and predict infrequent class categories. 3DSSN obtained 84.53 mAP 3D on the KITTI car category and 73.4 NDS on the nuScenes dataset, outperforming previous state of the art. Further, the result of FS-3DSSN on nuScenes is also encouraging for detecting infrequent categories while maintaining accuracy in frequent classes. [ABSTRACT FROM AUTHOR]

Pose estimation is an important component of many real-world computer vision systems. Most existing pose estimation algorithms need a large number of point correspondences to accurately determine the pose of an object. Since the number of point correspondences depends on the object’s appearance, lighting and other external conditions, detecting many points may not be feasible. In many real-world applications, the movement of objects is limited, e.g. due to gravity. Hence, detecting objects with only three degrees of freedom is usually sufficient. This allows us to improve the accuracy of pose estimation by changing the underlying equations of the perspective-n-point problem to three variables instead of six. By using the simplified equations, our algorithm is more robust against detection errors with limited point correspondences. In this article, we study three scenarios where such constraints apply. The first one is about parking a vehicle on a specific spot. Here, a stationary camera is detecting the vehicle to assist the driver. The second scenario describes the perspective of a moving camera detecting objects in its environment. This scenario is common for driver assistance systems, autonomous cars or mobile robots. Third, we describe a camera observing objects from a birds-eye view, which occurs in industrial applications. In all three scenarios, observed objects can only move in the ground plane and rotate around the vertical axis. Hence, three degrees of freedom are sufficient to estimate the pose. Experiments with synthetic data and real-world photographs have shown that our algorithm outperforms state-of-the-art pose estimation algorithms. Depending on the scenario, our algorithm is able to achieve 50% better accuracy, while being equally fast. [ABSTRACT FROM AUTHOR]

Object detection in road scenarios is crucial for intelligent transport systems and autonomous driving, but complex traffic conditions pose significant challenges. This paper introduces Z-You Only Look Once version 8 small (Z-YOLOv8s), designed to improve both accuracy and real-time efficiency under real-world uncertainties. By incorporating Revisiting Perspective Vision Transformer (RepViT) and C2f into the YOLOv8s framework, and integrating the Large Selective Kernel Network (LSKNet), the model enhances spatial feature extraction. Additionally, the YOLOv8s backbone is optimized with Space-to-Depth Convolution (SPD-Conv) for better small object detection. The Softpool-Spatial Pyramid Pooling Fast (SoftPool-SPPF) module ensures precise characteristic information preservation. Z-YOLOv8s improves mean average precision (mAP)@0.5 on the Berkeley Deep Drive 100 K (BDD100K) and Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) datasets by 7.3 % and 3.8 %, respectively. It also achieves accuracy increases of 5.7 % and 6.5 % in Average Precision (AP)-Small, and a real-time detection speed of 78.41 frames per second (FPS) on the BDD100K. Z-YOLOv8s balances detection precision and processing speed more effectively than other detectors, as demonstrated by experimental results and comparisons. • Designing a RepViTC2f module, to address occluded object challenges in traffic scenarios. • SoftPool-SPPF enhances small object detection by refining edge details. • Z-YOLOv8s balances speed and precision on BDD100K and KITTI datasets. [ABSTRACT FROM AUTHOR]