Complex Urban Scenarios Research Articles

Advancing complex urban traffic forecasting: A fully attentional spatial-temporal network enhanced by graph representation

Accurate urban traffic forecasting is essential for intelligent transportation systems (ITS). However, the majority of existing forecasting methodologies predominantly concentrate on point-based forecasts (e.g., traffic detector forecasts). A limited number of them pay attention to the urban bidirectional road segments and the complex road network topology. To advance accurate traffic forecasting in complex urban scenarios, this paper proposes a Graph Representation enhanced Fully Attentional Spatial-Temporal network (GR-FAST). First, we construct a refined bidirectional road network graph (BRG) to depict the urban road network topology more accurately, particularly focusing on the turning patterns at intersections. Then, we adopt the graph representation methodology and introduce spatial information encoding (SIE) to explicitly characterize the significance of roads and network structure from multiple perspectives. Enhanced by SIE, spatial attention can capture spatial dependencies from both road network topologies and traffic pattern similarities, thereby forming a unified urban spatial cognition. Finally, a multi-scale residual perception (MRP) module is designed to balance the interplay of short-term temporal variability and long-term periodicity. Experiments on a real-world urban dataset from Wuhan, China, demonstrate that GR-FAST outperforms the state-of-the-art deep learning methods, achieving an improvement of 9.19%. Furthermore, ablation studies suggest that the explicit incorporation of complex road spatial topologies can significantly enhance forecasting accuracy.

Research on the strategy of cruise control system for urban traffic jam assistant

This paper proposes a hierarchical cruise control strategy that can adapt to urban traffic scenarios. Firstly, the perception layer defines a target selection strategy by using a novel trapezoidal dangerous area. Then, the upper controller contains speed follow mode and distance follow mode. An improved PI mode with integral separation and saturation terms is developed to achieve the desired speed following control, a safe distance mode via variable time headway is developed to facilitate the safe and smooth distance following control. The lower controller is designed based on feedforward and PI feedback. Finally, the experimental results demonstrate that the strategy can effectively select suitable target vehicles in complex urban scenarios. With the improved safe distance model and the third-order Bessel actuator response delay model, the strategy exhibits adaptability and stability. This study offers valuable suggestions and guidance for designing and applying the cruise control strategy for urban traffic scenarios.

Improving real-time object detection in Internet-of-Things smart city traffic with YOLOv8-DSAF method

With the rise of global smart city construction, target detection technology plays a crucial role in optimizing urban functions and improving the quality of life. However, existing target detection technologies still have shortcomings in terms of accuracy, real-time performance, and adaptability. To address this challenge, this study proposes an innovative target detection model. Our model adopts the structure of YOLOv8-DSAF, comprising three key modules: depthwise separable convolution (DSConv), dual-path attention gate module (DPAG), and feature enhancement module (FEM). Firstly, DSConv technology optimizes computational complexity, enabling real-time target detection within limited hardware resources. Secondly, the DPAG module introduces a dual-channel attention mechanism, allowing the model to selectively focus on crucial areas, thereby improving detection accuracy in high-dynamic traffic scenarios. Finally, the FEM module highlights crucial features to prevent their loss, further enhancing detection accuracy. Additionally, we propose an Internet of Things smart city framework consisting of four main layers: the application domain, the Internet of Things infrastructure layer, the edge layer, and the cloud layer. The proposed algorithm utilizes the Internet of Things infrastructure layer, edge layer, and cloud layer to collect and process data in real-time, achieving faster response times. Experimental results on the KITTI V and Cityscapes datasets indicate that our model outperforms the YOLOv8 model. This suggests that in complex urban traffic scenarios, our model exhibits superior performance with higher detection accuracy and adaptability. We believe that this innovative model will significantly propel the development of smart cities and advance target detection technology.

An effective object detection algorithm for UAV-based urban regulation

Target detection from the perspective of UAV has great potential in the field of urban regulation, limited by the dense small targets, severe environmental obstructions, camera shake, and changes in lighting conditions in the aerial view of drones, the existing object detection algorithms cannot effectively undertake this task. This paper introduces two lightweight feature extraction modules based on YOLOv5, which are C3-Faster with PConv and COT3 with transformer structure. Meanwhile, an extra small detection head is added to the output layer. These approaches enhance accuracy while maintaining the advantages of being lightweight and easy to deploy. The ablation experiments and comparative experiments are designed to verify the effectiveness of these modules. The algorithm presented in this paper can be deployed into embedded systems of small UAVs to assist UAVs in completing various regulatory tasks in complex urban scenarios.

A physics-informed deep learning-based urban building thermal comfort modeling and prediction framework for identifying thermally vulnerable building stock

PurposeDue to the increasing frequency of extreme weather and densifying urban landscapes, residences are susceptible to heat-related discomfort, especially those in a naturally ventilated built environment in tropical climates. Indoor thermal comfort is thus paramount to building sustainability and improving occupants' health and well-being. However, to assess indoor thermal comfort considering the urban context, it is conventional to use questionnaire surveys and monitoring units, which are both case-centric and time-intensive. This study presents a dynamic computational thermal comfort modeling framework that can determine indoor thermal comfort at an urban scale to bridge this gap.Design/methodology/approachThe framework culminates in developing a deep learning model for predicting the accurate hourly indoor temperature of urban building stock by the coupling urban scale capabilities of environment modeling with single-building dynamic thermal simulations.FindingsUsing the framework, a surrogate model is created and verified for Dharavi, India's informal urban settlement. The results indicated that the developed surrogate model could predict the building's indoor temperature in several complex new urban scenarios with different building orientations, layouts, building-to-building distances and surrounding building heights, using five different random urban representative scenarios as the training set. The prediction accuracy was reliable, as evidenced by the mean bias error (MBE) and coefficient of (CV) root mean squared error (MSE) falling between 0 and 5%. The findings also showed that if the urban context is ignored, estimates of annual discomfort hours may be inaccurate by as much as 70%.Social implicationsThe developed computational framework could help regulators and policymakers engage in more informed and quantitative decision-making and direct efforts to enhance the thermal comfort of low-income dwellings and informal settlements.Originality/valueUp to this point, majority of literature that has been presented has concentrated on building a body of knowledge about urban-based modeling from an energy management standpoint. In contrast, this study suggests a dynamic computational thermal comfort modeling framework that takes into account the urban context of the neighborhood while examining the indoor thermal comfort of the residential building stock.

Towards high‐definition vector map construction based on multi‐sensor integration for intelligent vehicles: Systems and error quantification

AbstractA lightweight, high‐definition vector map (HDVM) enables fully autonomous vehicles. However, the generation of HDVM remains a challenging problem, especially in complex urban scenarios. Moreover, numerous factors in the urban environment can degrade the accuracy of HDVM, necessitating a reliable error quantification. To address these challenges, this paper presents an open‐source and generic HDVM generation pipeline that integrates the global navigation satellite system (GNSS), inertial navigation system (INS), light detection and ranging (LiDAR), and camera. The pipeline begins by extracting semantic information from raw images using the Swin Transformer. The absolute 3D information of semantic objects is then retrieved using depth from the 3D LiDAR, and pose estimation from GNSS/INS integrated navigation system. Vector information (VI), such as lane lines, is extracted from the semantic information to construct the HDVM. To assess the potential error of the HDVM, this paper systematically quantifies the impacts of two key error sources, segmentation and LiDAR‐camera extrinsic parameter error. An error propagation scheme is first formed to illustrate how these errors fundamentally influence the accuracy of the HDVM. The effectiveness of the proposed pipeline is demonstrated through our codeavailable at https://github.com/ebhrz/HDMap. The performance is verified using typical datasets, including indoor garages and complex urban scenarios.

Global Navigation Satellite System/Inertial Measurement Unit/Camera/HD Map Integrated Localization for Autonomous Vehicles in Challenging Urban Tunnel Scenarios

Lane-level localization is critical for autonomous vehicles (AVs). However, complex urban scenarios, particularly tunnels, pose significant challenges to AVs’ localization systems. In this paper, we propose a fusion localization method that integrates multiple mass-production sensors, including Global Navigation Satellite Systems (GNSSs), Inertial Measurement Units (IMUs), cameras, and high-definition (HD) maps. Firstly, we use a novel electronic horizon module to assess GNSS integrity and concurrently load the HD map data surrounding the AVs. This map data are then transformed into a visual space to match the corresponding lane lines captured by the on-board camera using an improved BiSeNet. Consequently, the matched HD map data are used to correct our localization algorithm, which is driven by an extended Kalman filter that integrates multiple sources of information, encompassing a GNSS, IMU, speedometer, camera, and HD maps. Our system is designed with redundancy to handle challenging city tunnel scenarios. To evaluate the proposed system, real-world experiments were conducted on a 36-kilometer city route that includes nine consecutive tunnels, totaling near 13 km and accounting for 35% of the entire route. The experimental results reveal that 99% of lateral localization errors are less than 0.29 m, and 90% of longitudinal localization errors are less than 3.25 m, ensuring reliable lane-level localization for AVs in challenging urban tunnel scenarios.

Achieving accurate trajectory predicting and tracking for autonomous vehicles via reinforcement learning-assisted control approaches

In complex urban traffic scenarios, autonomous vehicles face significant challenges in adapting to diverse and dynamic traffic conditions. Reward-based reinforcement learning has emerged as an effective approach to tackle these challenges. This paper presents a novel method that combines deep reinforcement learning with automotive dynamics systems. Building upon the Double Deep Q-learning algorithm, our approach integrates a Recurrent Neural Network with Gated Recurrent Units to enhance the environmental exploration capabilities of autonomous vehicles. To obtain more precise reward values, we introduce a trajectory tracking algorithm based on a combination of proportional-integral-derivative control and feedforward control within the automotive dynamics system. The proportional-integral-derivative controller is utilized for longitudinal control, while the Error-Optimized feedforward controller enhances lateral control, thereby improving trajectory tracking accuracy. Finally, extensive simulation experiments are conducted to evaluate the proposed method, comparing it against other baseline methods in terms of vehicle following and lane-changing scenarios. The results demonstrate that our approach significantly improves both the reward values and control performance of the algorithm.

Collaborative gas source localization strategy with networked nano-drones in unknown cluttered environments

This paper introduces a novel approach for improving gas source localization in dynamic urban environments, employing a swarm of nano-Crazyflie drones through a hybrid strategy that integrates Adaptive Robotic Particle Swarm Optimization (ARPSO) with Bidirectional Brain Emotional Learning (BBEL). The proposed method refines the ARPSO algorithm into two phases: an initial exploration phase and a subsequent seeking phase. The exploration phase activates when no gas is detected, seamlessly transitioning to the seeking phase upon gas detection. This facilitates efficient information exchange and desired velocity generation within the particle swarm. During the seeking phase, particles conduct measurements, share the global best position, intelligently navigate obstacles, and avoid collisions while directly identifying the global concentration field maximum. Unlike conventional methods, the ARPSO algorithm autonomously adapts its parameters online, mitigating algorithmic failures and local optima challenges. It also considers the limited communication of robots in complex environments, a factor often overlooked in recent methods. To enhance plume tracking robustness, facilitate the controller gain tuning process, and ensure system resilience against environmental disturbances and uncertainties, the BBEL method incorporates a sophisticated fuzzy neural network structure with reinforcement learning-based adaptation mechanisms. The stability of the closed-loop control system is rigorously proven using Lyapunov theory. Numerical simulations in complex urban scenarios validate the algorithm’s effectiveness, showcasing a significant 20% improvement in turnaround time and a flawless 100% success rate with no collisions, in addition to enhanced capabilities in handling uncertainties and disturbances compared to benchmarks like ARPSO-BBEL with unlimited communication range and no delayed communications (ARPSO-BBELU), Sniffy Bug (SB)-BBELU, and ARPSO-PIDU.

Real-Time 3D Routing Optimization for Unmanned Aerial Vehicle using Machine Learning

In the realm of Unmanned Aerial Vehicles (UAVs) for civilian applications, the surge in demand has underscored the need for sophisticated technologies. The integration of Unmanned Aerial Systems (UAS) with Artificial Intelligence (AI) has become paramount to address challenges in urban environments, particularly those involving obstacle collision risks. These UAVs are equipped with advanced sensor arrays, incorporating LiDAR and computer vision technologies. The AI algorithm undergoes comprehensive training on an embedded machine, fostering the development of a robust spatial perception model. This model enables the UAV to interpret and navigate through the intricate urban landscape with a human-like understanding of its surroundings. During mission execution, the AI-driven perception system detects and localizes objects, ensuring real-time awareness. This study proposes an innovative real-time three-dimensional (3D) path planner designed to optimize UAV trajectories through obstacle-laden environments. The path planner leverages a heuristic A* algorithm, a widely recognized search algorithm in artificial intelligence. A distinguishing feature of this proposed path planner is its ability to operate without the need to store frontier nodes in memory, diverging from conventional A* implementations. Instead, it relies on relative object positions obtained from the perception system, employing advanced techniques in simultaneous localization and mapping (SLAM). This approach ensures the generation of collision-free paths, enhancing the UAV's navigational efficiency. Moreover, the proposed path planner undergoes rigorous validation through Software-In-The-Loop (SITL) simulations in constrained environments, leveraging high-fidelity UAV dynamics models. Preliminary real flight tests are conducted to assess the real-world applicability of the system, considering factors such as wind disturbances and dynamic obstacles. The results showcase the path planner's effectiveness in providing swift and accurate guidance, thereby establishing its viability for real-time UAV missions in complex urban scenarios.

Deep Reinforcement Learning Based Integrated Eco-Driving Strategy for Connected and Automated Electric Vehicles in Complex Urban Scenarios

Deep Reinforcement Learning Based Integrated Eco-Driving Strategy for Connected and Automated Electric Vehicles in Complex Urban Scenarios

Safety Performance Assessment via Virtual Simulation of V2X Warning Triggers to Cyclists with Models Created from Real-World Testing

The overall crash statistics in the EU still show a very significant number of car–cyclist crashes. Within the Horizon 2020 project Safe-Up, countermeasures have been developed to reduce this number. One of these countermeasures involves a V2X-enhanced on-board unit for cycles, which can provide on-time warning triggers. The research assumption was based on studying the benefits of connectivity in enhancing cyclists’ safety. This study assessed the performance of this potential technology both qualitatively by analyzing volunteer feedback during physical testing and quantitatively by virtual simulations. The volunteers’ study showed positive findings on system’s safety relevance, user experience, and user acceptance. The method applied for the virtual simulation is a prospective safety performance assessment with reconstructed accident scenarios based on the GIDAS database and cyclist behavior models, obtained from physical testing. The results using a warning trigger 4 s prior to the collision showed a potential safety benefit of approximately 98%. It should be noted that this trigger time was found to be quite early in both physical testing and virtual simulation. Further research is required to evaluate the system’s performance in more complex urban scenarios, as well as to design the human–machine interaction strategies for optimal accident avoidance.

Research on Intelligent Vehicle Operation Risk Assessment and Early Warning Based on Predictive Risk Field

In order to enhance the driving safety of intelligent vehicles in complex road scenarios, a method for vehicle operation risk assessment and early warning based on the predictive risk field is proposed. The temporal feature vector composed of the spatiotemporal state characteristics of the ego vehicle and surrounding traffic participants is taken as input data for the Attention‐Bidirectional Long‐Short Term Memory (Attention‐BiLSTM) model, which is trained to establish the desired mapping relationship. By predicting the motion state of the target vehicle and utilizing an improved risk field model based on the target vehicle of heading angle, the predictive risk field is obtained. This allows for the assessment of the ego vehicle operational risks. The risk warning model is integrated to provide risk early warning, and the safety path for the ego vehicle is planned based on the interaction between the predictive risk field equipotential lines and the cubic spline curves. Experimental results demonstrate that the proposed vehicle operation risk assessment and early warning model is effective in providing early warnings and safe path references for the ego vehicle in complex urban road test scenarios.

A Hierarchical Vehicle Behavior Prediction Framework With Traffic Signals and Interactive Agents

Vehicle behavior prediction in complex urban scenarios with traffic signals and interactive agents is an important yet complicated task for autonomous vehicles (AVs). In this work, a hierarchical vehicle behavior prediction framework is proposed to incorporate the traffic signal information and model the interaction between vehicles. The framework predicts vehicle behaviors in two stages, discrete intention prediction and continuous trajectory prediction. In the discrete intention prediction stage, Bayesian network is adopted to provide a high-level behavior prediction of the principle other vehicle. The discrete prediction results are forwarded to the second stage, where a continuous trajectory is predicted with maximum entropy inverse reinforcement learning and potential game. The framework is designed to be able to capture the difference among human drivers with parameterized driver characteristics. The proposed predictor is validated in two scenarios: the yellow light running scenario and the right-turn scenario. The trajectory prediction average displacement error of the yellow light running scenario is 0.695m for a 3-second prediction interval, and the prediction accuracy of the right-turn vehicle in the right-turn scenario is 0.51m for a 2-second prediction interval.

Reliability optimization in IRS-assisted UAV networks

Unmanned aerial vehicles (UAVs) and intelligent reflecting surfaces (IRSs) are regarded as critical technologies which need to be developed for the sixth-generation mobile network (6G). In this paper, we study the reliability of the IRS-assisted Unmanned aerial vehicles (UAV) communication system with limited UAV energy in complex urban scenarios. We present the problem by joint optimization of the IRS's scheduling, the UAV's trajectory, the IRS's phase shift, and the UAV's transmit power for maximizing the reliability of the transmission link under the constraints of UAV energy and channel error rate. The concavity of the problem is difficult to judge, and the problem is hard to find a substitution function to convert it into a tractable form. Therefore, we design a chaotic adaptative hybrid whale optimization algorithm (CAHWOA) to address the problem. CAHWOA is implemented by using alternately the chaotic adaptation whale optimization algorithm (CAWOA) and the variational perturbation binary whale optimization algorithm (VPBWOA). CAWOA and VPBWOA are used for solving the continuous variable subproblem and the binary variable subproblem, respectively. The numerical results show that the joint optimization of IRS and UAV improves the transmission link reliability by nearly 30% compared with the two baseline schemes. CAHWOA can improve the convergence rate by about 18% compared with the baseline algorithms. Meanwhile, it can enhance the optimization-seeking accuracy by about 0.04.

Multi-Tier 3D Trajectory Planning for Cellular-Connected UAVs in Complex Urban Environments

Cellular-connected unmanned aerial vehicles (UAVs) present a viable solution to address communication and navigation limitations by leveraging base stations for air–ground communication. However, in complex urban scenarios with stringent communication requirements, achieving asymmetrical control becomes crucial to strike a balance between communication reliability and flight safety. Moreover, existing cellular-connected UAV trajectory planning algorithms often struggle to handle real scenes with sudden and intricate obstacles. To address the aforementioned challenges, this paper presents the multi-tier trajectory planning method (MTTP), which takes into account air–ground communication service assurance and collision avoidance in intricate urban environments. The proposed approach establishes a flight risk model that accounts for both the outage probability of UAV-ground base station (GBS) communication and the complexity of flight environments, and transforms the inherently complex three-dimensional (3D) trajectory optimization problem into a risk distance minimization model. To optimize the flight trajectory, a hierarchical progressive solution approach is proposed, which combines the strengths of the heuristic search algorithm (HSA) and deep reinforcement learning (DRL) algorithm. This innovative fusion of techniques empowers MTTP to efficiently navigate complex scenarios with sudden obstacles and communication challenges. Simulations show that the proposed MTTP method achieves a more superior performance of trajectory planning than the conventional communication-based solution, which yields a substantial reduction in flight distance of at least 8.49% and an impressive 10% increase in the mission success rate. Furthermore, a real-world scenario is chosen from the Yuhang District, Hangzhou (a southern Chinese city), to validate the practical applicability of the MTTP method in highly complex operating scenarios.

Overcoming driving challenges in complex urban traffic: A multi-objective eco-driving strategy via safety model based reinforcement learning

This study proposes a novel eco-driving control strategy for connected and automated hybrid electric vehicles, which utilizes deep reinforcement learning (DRL) to optimize various aspects of driving performance, including fuel economy, ride comfort, and travel efficiency, in complex urban traffic scenarios. The proposed strategy incorporates a driving safety model that predicts potential risk associated with the DRL agent's planned speed, thus ensuring the safety of the DRL based eco-driving strategy. Additionally, we propose a multi-objective composite reward function design scheme that considers various constraints caused by traffic elements, such as traffic lights, preceding vehicles, road curvature, and speed limit. This design scheme enables the proposed strategy to effectively adapt to diverse driving challenges in complex urban traffic scenarios. To evaluate the proposed strategy, we develop an urban traffic simulation model based on real-world road and traffic data from Shanghai, China. This model is used as the test scenario and can reflect real urban traffic conditions. The simulation results demonstrate the capability of the proposed strategy to safely and efficiently control vehicles to complete driving tasks in complex urban scenarios. Moreover, the proposed strategy excels in simultaneously optimizing the driving comfort and fuel consumption of the controlled vehicle.

Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction

In complex and dynamic urban traffic scenarios, the accurate trajectory prediction of surrounding pedestrians with interactive behaviors plays a vital role in the self-driving system. Intrinsic factors and extrinsic factors will inevitably influence the pedestrians trajectory. Intrinsic factors such as pedestrians diversified intentions bring rich and diverse multi-modal future possibilities. Besides, extrinsic factors affecting the future trajectory are accompanied by context semantics such as interactions among pedestrians. However, most of the existing methods discuss two problems (interaction and intention) separately. Considering both two factors impact the trajectory of pedestrians, a Triple Policies Fused Hierarchical Graph Neural Networks (Tri-HGNN) is proposed to model spatial and temporal interactions and intentions among the whole scene of pedestrians at each time step and predict the multiple future trajectories. Tri-HGNN contains three different policies: (i) Extrinsic-level policy is used to extract spatial nodes embedding from the interaction graph of pedestrian trajectories by using the Graph Attention Network. (ii) Intrinsic-level policy adopts the Graph Convolutional Network to infer the human intention for more accurate prediction. Moreover, human intention is influenced by the intrinsic interaction generated among pedestrians, so we fuse the interaction features to grasp the influence of the extrinsic interaction. (iii) Basic-level policy then integrates the heuristic information obtained from other two policies and concatenates it with historical trajectories to make multiple predictions through Temporal Convolution Network. Experimental results show that our model improves performance compared with state-of-the-art methods on the ETH/UCY and SDD benchmarks.

Deep reinforcement learning for optimal rescue path planning in uncertain and complex urban pluvial flood scenarios

An urban pluvial flood is a devastating, costly natural disaster requiring effective rescue path planning to mitigate the loss of lives and property. The inherent uncertainty and complexity of the risks associated with urban flooding limit the ability to plan optimal rescue paths that prioritize both timeliness and safety. This study addresses the challenge by proposing an innovative assessment methodology to output risk values and probability representing safety and timeliness in each passable area while simulating real-world flood scenarios. Furthermore, the paper develops a pioneering path-planning algorithm based on deep reinforcement learning, incorporating improved stochastic reward exploitation and heterogeneous reward exploration mechanisms to function in a simulation rescue path-planning scenario with uncertainty and complexity. According to the findings, the proposed algorithm outperforms current state-of-the-art algorithms in converging to the optimal path, fully sampling, and running efficiency. The study contributes to theoretical progress on urban pluvial flood rescue, deep reinforcement learning, risk assessment, and decision intelligence while offering practical implications for smart cities, emergency management, and optimizing real-world problems by employing artificial intelligence.

An ROI Optimization Method Based on Dynamic Estimation Adjustment Model

An important research direction in the field of traffic light recognition of autonomous systems is to accurately obtain the region of interest (ROI) of the image through the multi-sensor assisted method. Dynamic evaluation of the performance of the multi-sensor (GNSS, IMU, and odometer) fusion positioning system to obtain the optimum size of the ROI is essential for further improvement of recognition accuracy. In this paper, we propose a dynamic estimation adjustment (DEA) model construction method to optimize the ROI. First, according to the residual variance of the integrated navigation system and the vehicle velocity, we divide the innovation into an approximate Gaussian fitting region (AGFR) and a Gaussian convergence region (GCR) and estimate them using variational Bayesian gated recurrent unit (VBGRU) networks and a Gaussian mixture model (GMM), respectively, to obtain the GNSS measurement uncertainty. Then, the relationship between the GNSS measurement uncertainty and the multi-sensor aided ROI acquisition error is deduced and analyzed in detail. Further, we build a dynamic estimation adjustment model to convert the innovation of the multi-sensor integrated navigation system into the optimal ROI size of the traffic lights online. Finally, we use the YOLOv4 model to detect and recognize the traffic lights in the ROI. Based on laboratory simulation and real road tests, we verify the performance of the DEA model. The experimental results show that the proposed algorithm is more suitable for the application of autonomous vehicles in complex urban road scenarios than the existing achievements.