LiDAR-based bird’s-eye-view (BEV) perception has emerged as an appealing approach for practical autonomous driving applications due to its direct leveraging of precise 3D structures and delivering efficient performance. This paradigm aims to jointly determine the semantics and motion states of various traffic participants on BEV grids. However, most existing LiDAR-based BEV perception methods primarily focus on motion prediction, leading to inferior semantic performance. To address this limitation, we propose a novel multi-frame, multi-view, and multi-task unified framework in this work, which enhances scene perception for both improved BEV semantic segmentation and comparative motion prediction performances. Our framework, named temporal range-point-voxel fusion (T-RPVFusion), leverages a sequence of LiDAR sweeps as input and jointly outputs semantic and motion information on BEV grids. In T-RPVFusion, we first introduce a novel multi-view semantic encoder that extracts high-quality semantic features from each LiDAR sweep. These semantic feature maps are then aggregated into an integrated feature map using the proposed bi-layer spatio-temporal pyramid network. Subsequently, the integrated feature map undergoes processing in both the semantic and motion heads and yields corresponding outputs, respectively. Extensive experiments conducted on Waymo and nuScenes show that our method outperforms previous state-of-the-art (SOTA) in terms of BEV semantic segmentation, while concurrently demonstrating comparable performance in motion prediction. Notably, our method achieves a significant improvement on BEV semantic segmentation task, attaining a mIOU of 49.5%, surpassing the previous SOTA with a great margin of + 12.1% mIOU on Waymo Open Dataset. The code is available athttps://github.com/thuwyl/trpvfusion

In autonomous driving, accurately identifying traffic participants that may influence vehicle behavior is crucial for effective system planning. To address this challenge, we propose a fault-tolerant mechanism for detecting action-induced objects, which significantly improves decision-making performance and system explainability. Since these objects are often linked to the vehicle’s driving intentions, we introduce a top-down attention network that adjusts attention weights for traffic participants based on navigational information. Additionally, we define potentially hazardous objects in the driving environment and employ supervised training with a classification head to detect them. To further enhance detection accuracy, we integrate a fault-tolerant process that merges attention maps with classification results, effectively reducing false positives and false negatives in identifying action-induced objects. Extensive testing validates the robustness and effectiveness of our approach, demonstrating its ability to improve both planning and interpretability in autonomous vehicles.

Reliable spatial and motion perception is essential for safe autonomous navigation. Recently, class-agnostic motion prediction on bird's-eye view (BEV) cell grids derived from LiDAR point clouds has gained significant attention. However, existing frameworks typically perform cell classification and motion prediction on a per-pixel basis, neglecting important motion field priors such as rigidity constraints, temporal consistency, and future interactions between agents. These limitations lead to degraded performance, particularly in sparse and distant regions. To address these challenges, we introduce \textbf{PriorMotion}, an innovative generative framework designed for class-agnostic motion prediction that integrates essential motion priors by modeling them as distributions within a structured latent space. Specifically, our method captures structured motion priors using raster-vector representations and employs a variational autoencoder with distinct dynamic and static components to learn future motion distributions in the latent space. Experiments on the nuScenes dataset demonstrate that \textbf{PriorMotion} outperforms state-of-the-art methods across both traditional metrics and our newly proposed evaluation criteria. Notably, we achieve improvements of approximately 15.24\% in accuracy for fast-moving objects, an 3.59\% increase in generalization, a reduction of 0.0163 in motion stability, and a 31.52\% reduction in prediction errors in distant regions. Further validation on FMCW LiDAR sensors confirms the robustness of our approach.

Accurate localization plays an important role in high-level autonomous driving systems. Conventional map matching-based localization methods solve the poses by explicitly matching map elements with sensor observations, generally sensitive to perception noise, therefore requiring costly hyperparameter tuning. In this paper, we propose an end-to-end localization neural network which directly estimates vehicle poses from surrounding images, without explicitly matching perception results with HD maps. To ensure efficiency and interpretability, a decoupled BEV neural matching-based pose solver is proposed, which estimates poses in a differentiable sampling-based matching module. Moreover, the sampling space is hugely reduced by decoupling the feature representation affected by each DoF of poses. The experimental results demonstrate that the proposed network is capable of performing decimeter level localization with mean absolute errors of 0.19m, 0.13m and 0.39° in longitudinal, lateral position and yaw angle while exhibiting a 68.8% reduction in inference memory usage.

Accurate spatial and motion understanding is critical for autonomous driving systems. While object-level perception models excel in structured environments, they struggle with open-set categories and often lack precise geometric representation. Occupancy-based, class-agnostic methods offer better scene expressiveness but typically ignore inter-agent interactions and fail to ensure physical consistency in motion predictions, limiting their reliability in complex traffic scenarios. In this paper, we propose LEGO-Motion, a novel class-agnostic motion prediction framework that bridges the gap between instance-level reasoning and occupancy-based modeling. Unlike conventional grid-based methods that treat each cell independently, LEGO-Motion introduces two key components: (1) the Interaction-Augmented Instance Encoder (IaIE), which models interactions among dynamic agents via cross-attention, and (2) the Instance-Enhanced BEV Encoder (IeBE), which improves motion consistency across instances through multi-stage feature fusion. These components enable our model to learn semantically coherent and physically plausible motion fields. Extensive experiments on the nuScenes dataset show that LEGO-Motion achieves a around 6% improvement in motion prediction accuracy over the previous state-of-the-art, while maintaining real-time inference at 21ms. Moreover, our method demonstrates strong generalization on a proprietary FMCW LiDAR benchmark. These results validate LEGO-Motion's effectiveness in capturing both global scene structure and fine-grained motion dynamics, making it a promising foundation for next-generation perception systems.

Pedestrian trajectory prediction is significant for autonomous vehicles, but the difficulty of pedestrian trajectory prediction lies in the accurate modeling of pedestrian multiple interactions. In this paper, we attempt to explore the essential features of pedestrian interaction and propose a pedestrian trajectory prediction method based on multiple interactions. Firstly, considering that the interaction between self-driving cars and pedestrians resembles a dynamic game process involving sequential adaptation, we map them to the same feature space and design a temporal cross-attention mechanism to model the interaction between pedestrians and vehicles. Meanwhile, pedestrian-scene interaction is affected by the global environment as well as the local environment. To capture the global information while preserving the spatial location of pedestrians in the scene, we design a pedestrian-scene heatmap fusion (PSHF) framework to model the pedestrian-scene interaction features. We validate the effectiveness of our algorithm on the publicly available JAAD and PIE datasets, achieving better performance than existing representative methods in both single-trajectory and multi-trajectory prediction tasks. We conducted a thorough ablation study, cross-dataset validation, and qualitative visualization experiments, demonstrating the effectiveness and robustness of our method.

Despite the remarkable advancements in autonomous driving, the challenge persists in achieving interpretable action decision-making, primarily owing to the intricate and ambiguous relationship between detected agents and driving intention. In this study, we introduce an interpretable action prediction model, denoted as the Prediction-Driven Attention Network (PDANet), designed to undertake action decisions and provide corresponding interpretations cohesively. The PDANet is inspired by the perceptual mechanisms inherent in human drivers, who allocate attention according to their driving intentions. Specifically, we elaborate a prediction module to generate vehicle prospective trajectories to characterize driving intentions. Subsequently, the features of this predicted trajectory are utilized to modulate the attention distribution among agents through the top-down attention module, yielding an attention map. Finally, two distinct task tokens are applied to aggregate agent features and generate the final output according to the derived attention map. Extensive experiments conducted on the publicly available BDD-OIA and nu-AR datasets demonstrate that our proposed method outperforms all prior works in terms of both action prediction and behavior interpretation tasks. Remarkably, our method attains a noteworthy enhancement in the behavior interpretation task, surpassing the previous state-of-the-art by a substantial margin of +10.8% in terms of F1-score on the nu-AR dataset. We also validate our algorithm on Carla Town05 long in a closed-loop decision-making scenario, highlighting the generality and robustness of our approach. Furthermore, qualitative results show that the agents selected by our model are more closely aligned with human cognitive processes.

In the rapidly evolving landscape of autonomous vehicles and advanced navigation systems, the accuracy of high-definition maps and real-time updating has become paramount. However, in this progression, the security of map data has not received adequate attention, although the accuracy of the data can be easily altered when the system is breached. Thus, this paper introduces a novel approach to democratizing the process of high-definition map updates by leveraging consortium blockchain technology specifically designed for Proof of Presence and Reputation (POP-R) to safeguard the update process. Our proposed system leverages the presence and reputation of vehicles through infrastructure nodes to enhance the accuracy and reliability of HD map updates. We created a trusted ecosystem for maintaining high-definition maps, marked by a superior safety score across three scenarios compared to the standard proof of reputation technique. Additionally, it demonstrates high efficiency, achieving 12,000 transactions per second (TPS) for data queries and more than 2,500 TPS for data writing in our blockchain network. This efficiency proved our prowess in the lightweight computational power required, suitable for decentralized and crowdsourced-based systems. Through our POP-R framework, we lay the foundation for a new decentralized approach to the evolution of high-definition maps in the era of autonomous mobility.

Two-dimensional (2-D) direction of arrival (DOA) estimation with a circular array based on the difference co-array has attracted considerable attention in past years. In this paper, the difference co-array position set of a circular array with arbitrary sensor arrangement is derived, and condition under which the maximum number of virtual co-array sensors can be provided by a circular array is presented. Then, an augmented uniform circular array (AUCA) is proposed, providing the maximum number of DOFs for arbitrary number of physical sensors. Compressive sensing based focusing method for the one-dimensional case is extend to 2-D wideband DOA estimation, where focusing on the difference co-array is adopted for performance improvement. Simulations show that better performance can be achieved by our proposed AUCA.

In high-definition maps (HD maps), the road lane centerline and lane topology graph play essential roles in navigation, planning, and decision-making. Existing research focusing on extracting physical infrastructure, such as lane boundaries, has made significant progress. But lane centerline detection and topology reasoning still remains challenging due to the severe overlapping centerlines and complicated topology. To tackle these challenges, we introduce an automatic lane topology extraction method for HD maps, termed LaneDAG, which extracts vectorized centerlines and their topology from prebuilt lane lines and road boundaries in HD maps. It formulates centerline extraction as a set prediction problem and lane topology prediction as a directed acyclic graph (DAG) construction problem. A novel mechanism that fusing geometric and attention-based features in the DAG is proposed to model the topological relationship between centerlines. Experiments conducted on the Argoverse 2 dataset demonstrate the proposed method’s superior performance compared to existing methods, showcasing its capability to extract lane centerlines and topology in HD maps automatically.

Motion forecasting is an important component in autonomous driving systems. One of the most challenging problems in motion forecasting is interactive trajectory prediction, whose goal is to jointly forecasts the future trajectories of interacting agents. To this end, we present a large-scale interactive trajectory prediction dataset named INT2 for INTeractive trajectory prediction at INTersections. INT2 includes 612,000 scenes, each lasting 1 minute, containing up to 10,200 hours of data. The agent trajectories are auto-labeled by a high-performance offline temporal detection and fusion algorithm, whose quality is further inspected by human judges. Vectorized semantic maps and traffic light information are also included in INT2. Additionally, the dataset poses an interesting domain mismatch challenge. For each intersection, we treat rush-hour and non-rush-hour segments as different domains. We benchmark the best open-sourced interactive trajectory prediction method on INT2 and Waymo Open Motion, under in-domain and cross-domain settings. The dataset, code and models are publicly available at https://github.com/AIRDISCOVER/INT2.

The Qinghai–Tibet Plateau is rich in water resources with numerous lakes, rivers, and glaciers, and, as a source of many rivers in Central Asia, it is known as the Asian Water Tower. Under global climate change, it is critical to understand the current influencing factors on surface water area in this region. Although there are numerous studies on surface water mapping, they are still limited by temporal/spatial resolution and record length. Moreover, the complicated topographic condition makes it challenging to map the surface water accurately. Here, we proposed an automatic two-step annual surface water classification framework using long time-series Landsat images and topographic information based on the Google Earth Engine (GEE) platform. The results showed that the producer accuracy (PA) and user accuracy (UA) of the surface water map in the Qinghai–Tibet Plateau in 2020 were 99% and 90%, respectively, and the Kappa coefficient reached 0.87. Our dataset showed high consistency with high-resolution images, indicating that the proposed large-scale water mapping method has great application potential. Furthermore, a new annual surface water area dataset on the Qinghai–Tibet Plateau from 2000 to 2020 was generated, and its relationship with climate, vegetation, permafrost, and glacier factors was explored. We found that the mean surface water area was about 59 481 km2, and there was a significant increasing trend (=322 km2/year,$p < 0.01$) during 2000–2020 in the plateau. Greening, warming, and wetting climate conditions contributed to the increase of surface water area. Active layer thickness and permafrost types may be the most related to the decrease of surface water area. This study provides important information for ecological assessment and protection of the plateau and promotes the implementation of sustainable development goals related to surface water resources.

It is critical for autonomous vehicles to recognize traffic police gestures timely and accurately. During the movement of the vehicle, the collected traffic police scales change all the time, in addition, the frequency and amplitude of actions of different traffic police are different. First, we use gesture normalization to fix the traffic police actions at a unified scale and remove the influence of scale changes on traffic police gesture recognition. Meanwhile, a fully adaptive spatial-temporal graph convolution network (FA-STGCN) is proposed to recognize the actions with different amplitude and frequencies. The adaptive spatial graph network can dig the latent joints connection relation of the traffic police under different gestures, which weakens the amplitude impact on the action recognition. The adaptive temporal graph network is composed of the global temporal module and the local temporal module. The global temporal module can obtain the coarse-grained features of the traffic police gestures’ speed and then naturally use the coarse-grained features to guide the local temporal module to adaptively learn the fine-grained temporal features of the traffic police action. The adaptive spatial graph network and the temporal graph network are alternately stacked to finally output accurate traffic police gestures. We thoroughly evaluated our method through intensive experiments, the result shows that our method achieved the best results on public datasets. What’s more, we proofed the effectiveness of each module and verified our methods for moving vehicles for the first time, the performance present meets the vehicle’s practical requirements.

The artificial colony (ABC) algorithm shows a relatively powerful exploration search capability but is constrained by the curse of dimensionality, especially on nonseparable functions, where its convergence speed slows dramatically. In this article, based on an analysis of the difference between updating mechanisms that include both all-variable and one-variable updating mechanisms, we find that when equipped with the former strategy, the algorithm rapidly converges to an optimal region, while with the latter strategy, it searches the solution space thoroughly. To utilize multivariable and one-variable updating mechanisms on nonseparable and separable functions, respectively, we embed an improved linkage identification strategy into the ABC by detecting the linkage between variables more effectively. Then, we propose three common strategies for ABC to improve its performance. First, a new approach that considers the historic experiences of the population is proposed to balance exploration and exploitation. Second, a new strategy for initializing scout bees is used to reduce the number of function evaluations. Finally, the individual with the worst performance is updated with a defined probability on multiple dimensions instead of one dimension, causing it to follow the population steps on nonseparable functions. This article is the first to propose all these concepts, which could be adopted for other ABC variants. The effectiveness of our algorithm is validated through basic, CEC2010, CEC2013, and CEC2014 functions and real-world problems.

Recognizing traffic command gestures with high accuracy and quick response at a low computational cost is a requisite for driver assistance or autonomous driving. However, it has been understudied for a long time. Existing research takes advantage of increasing development in human action recognition but pays little attention to onboard conditions. In this article, we propose a simple but effective recognition model based on human upper-body geometric features and a long short-term memory (LSTM) network. The handcrafted geometric features can easily be calculated with estimated 2-D human keypoints at a low computational cost but are discriminative and sufficient in classification. Offline and online inferences are implemented to comprehensively evaluate the proposed model. For the sake of robustness required in the automotive domain, dual voting is designed to filter the output in online inference. On the recently published Chinese traffic police gesture (CTPG) dataset, the presented approach is the best with a remarkable improvement of approximately 8% compared to previous LSTM-based methods with handcrafted spatial features and is competitive with advanced GCN-based deep learning methods. The tradeoff pattern is explored to demonstrate how accuracy and response time alter with different training and inference strategies so that a balanced setup can be manually chosen under various application scenarios. Field tests are also carried out with an experimental vehicle, and the results uncover the present gap between research and practical application to some extent, moving a step closer to real-life traffic command gesture recognition.

Formal specifications of software systems need to evolve in many ways during system development. Not only are changes required to refine the specification towards an implementation, they are also required in response to changes in requirements, or to incorporate different aspects of the system, e.g., fault tolerance or timing, initially ignored in order to simplify reasoning. This paper presents an approach for evolving Z specifications by the step-wise application of a number of simple rules. These rules not only document the specification's evolution, but also make precise how safety properties of the system evolve with the specification. Hence, reasoning about these properties performed on the original specification need not be repeated on the new specification.

We present results from the second pilot project in the international Verification Grand Challenge: a formally verified specification of a POSIX-compliant file store using the Z/Eves theorem prover. The project's overall objective is to build a verified file store for space-flight missions. Our specification of the file store is based on Morgan & Sufrin's specification of the UNIX filing system; the proof and its mechanisation in Z/Eves are novel. We show how our work contributes towards building a verified software repository: a set of general theories and experiments reusable across different domains.

The protein-protein interaction networks, or interactome networks, have been shown to have dynamic modular structures, yet the functional connections between and among the modules are less well understood. Here, using a new pipeline to integrate the interactome and the transcriptome, we identified a pair of transcriptionally anticorrelated modules, each consisting of hundreds of genes in multicellular interactome networks across different individuals and populations. The two modules are associated with cellular proliferation and differentiation, respectively. The proliferation module is conserved among eukaryotic organisms, whereas the differentiation module is specific to multicellular organisms. Upon differentiation of various tissues and cell lines from different organisms, the expression of the proliferation module is more uniformly suppressed, while the differentiation module is upregulated in a tissue- and species-specific manner. Our results indicate that even at the tissue and organism levels, proliferation and differentiation modules may correspond to two alternative states of the molecular network and may reflect a universal symbiotic relationship in a multicellular organism. Our analyses further predict that the proteins mediating the interactions between these modules may serve as modulators at the proliferation/differentiation switch.

The assignment of orthologous genes between a pair of genomes is a fundamental and challenging problem in comparative genomics. Existing methods that assign orthologs based on the similarity between DNA or protein sequences may make erroneous assignments when sequence similarity does not clearly delineate the evolutionary relationship among genes of the same families. In this paper, we present a new approach to ortholog assignment that takes into account both sequence similarity and evolutionary events at a genome level, where orthologous genes are assumed to correspond to each other in the most parsimonious evolving scenario under genome rearrangement. First, the problem is formulated as that of computing the signed reversal distance with duplicates between the two genomes of interest. Then, the problem is decomposed into two new optimization problems, called minimum common partition and maximum cycle decomposition, for which efficient heuristic algorithms are given. Following this approach, we have implemented a high-throughput system for assigning orthologs on a genome scale, called SOAR, and tested it on both simulated data and real genome sequence data. Compared to a recent ortholog assignment method based entirely on homology search (called INPARANOID), SOAR shows a marginally better performance in terms of sensitivity on the real data set because it is able to identify several correct orthologous pairs that are missed by INPARANOID. The simulation results demonstrate that SOAR, in general, performs better than the iterated exemplar algorithm in terms of computing the reversal distance and assigning correct orthologs.