Complex Dynamic Environments Research Articles

Automatic Tracking Control Based on Bayesian Optimization

Abstract. This paper discusses the control problem of intelligent tracking vehicle in complex dynamic environment, and puts forward a solution combining Bayesian optimization algorithm and PID control. Traditional PID control faces many challenges when dealing with time-varying and nonlinear systems, but the operating environment of intelligent tracking vehicles is complex and changeable, and it is difficult to adjust parameters in real time to adapt to different operating conditions. Therefore, this paper introduces Bayesian optimization algorithm to predict the optimal parameter combination of PID controller by constructing hyperparameter optimization model, thus effectively improving the control performance and robustness of the system.

Enhancing YOLOv8’s Performance in Complex Traffic Scenarios: Optimization Design for Handling Long-Distance Dependencies and Complex Feature Relationships

In recent years, the field of deep learning and computer vision has increasingly focused on the problem of vehicle target detection, becoming the forefront of many technological innovations. YOLOv8, as an efficient vehicle target detection model, has achieved good results in many scenarios. However, when faced with complex traffic scenarios, such as occluded targets, small target detection, changes in lighting, and variable weather conditions, YOLOv8 still has insufficient detection accuracy and robustness. To address these issues, this paper delves into the optimization strategies of YOLOv8 in the field of vehicle target detection, focusing on the EMA module in the backbone part and replacing the original SPPF module with focal modulation technology, all of which effectively improved the model’s performance. At the same time, modifications to the head part were approached with caution to avoid unnecessary interference with the original design. The experiment used the UA-DETRAC dataset, which contains a variety of traffic scenarios, a rich variety of vehicle types, and complex dynamic environments, making it suitable for evaluating and validating the performance of traffic monitoring systems. The 5-fold cross-validation method was used to ensure the reliability and comprehensiveness of the evaluation results. The final results showed that the improved model’s precision rate increased from 0.859 to 0.961, the recall rate from 0.83 to 0.908, and the mAP50 from 0.881 to 0.962. Meanwhile, the optimized YOLOv8 model demonstrated strong robustness in terms of detection accuracy and the ability to adapt to complex environments.

Vision-based multi-view reconstruction for high-precision part positioning in industrial robot machining

Traditional industrial robot machining typically involves offline programming of machining trajectories, with part positioning accomplished through robotic tools. However, this method has limitations, particularly when dealing with complex workpieces and dynamic environments. To tackle this problem, this paper introduces a vision-based multi-view reconstruction approach for part positioning in industrial robot machining. Multiple cameras capture multi-view images of the part, reconstruct its 3D point cloud, and match it with the CAD model of the part to achieve high-precision positioning. This method effectively translates offline programmed robot trajectories to real-world scenarios, improving the flexibility and adaptability of robot machining. Experimental results confirm the high accuracy and robustness of the proposed vision-based reconstruction method, achieving a relative error of less than 2 mm for 85.7% of points, with an overall root mean square error of 1.09 mm.

3D Voxel-based Path Planning for AVs in Dynamic Complex Environments

Abstract. With the rapid advancement of science and technology, autonomous driving technology has been widely applied in various fields such as the automotive industry, intelligent warehousing, and emergency rescue. Traditional 2D path planning has limitations in dealing with the complex and dynamic environments of the real world, making it difficult to meet the high precision and complex path planning requirements of autonomous vehicles. With the development of data acquisition technology, large-scale point cloud data has become easily accessible, which can better extract the three-dimensional complex of the environment. Nevertheless, the unstructured nature and vast amount of point cloud data make path planning on it extremely challenging. Voxel models can effectively compress point cloud data and provide neighboring topological structures. In view of this, we propose a voxel-based framework for the path planning of autonomous vehicles in dynamic and complex 3D environments, which involves converting point cloud data into voxels and extracting navigation spaces. Subsequently, deep reinforcement learning techniques are utilized to achieve obstacle avoidance and navigation on the voxel map. It is expected that this framework will contribute to the development of 3D path planning and enhance the utilization of point cloud data in the navigation field.

Multi-Step Temperature Prognosis of Lithium-Ion Batteries for Real Electric Vehicles Based on a Novel Bidirectional Mamba Network and Sequence Adaptive Correlation

The battery systems of electric vehicles (EVs) are directly impacted by battery temperature in terms of thermal runaway and failure. However, uncertainty about thermal runaway, dynamic conditions, and a dearth of high-quality data sets make modeling and predicting nonlinear multiscale electrochemical systems challenging. In this work, a novel Mamba network architecture called BMPTtery (Bidirectional Mamba Predictive Battery Temperature Representation) is proposed to overcome these challenges. First, a two-step hybrid model of trajectory piecewise–polynomial regression and exponentially weighted moving average is created and used to an operational dataset of EVs in order to handle the problem of noisy and incomplete time-series data. Each time series is then individually labeled to learn the representation and adaptive correlation of the multivariate series to capture battery performance variations in complex dynamic operating environments. Next, a prediction method with multiple steps based on the bidirectional Mamba is suggested. When combined with a failure diagnosis approach, this scheme can accurately detect heat failures due to excessive temperature, rapid temperature rise, and significant temperature differences. The experimental results demonstrate that the technique can accurately detect battery failures on a dataset of real operational EVs and predict the battery temperature one minute ahead of time with an MRE of 0.273%.

Adaptive Deep Ant Colony Optimization–Asymmetric Strategy Network Twin Delayed Deep Deterministic Policy Gradient Algorithm: Path Planning for Mobile Robots in Dynamic Environments

Path planning is one of the main focal points and challenges in mobile robotics research. Traditional ant colony optimization (ACO) algorithms encounter issues such as low efficiency, slow convergence, and a tendency to become stuck in local optima and search stagnation when applied to complex dynamic environments. Addressing these challenges, this study introduces an adaptive deep ant colony optimization (ADACO) algorithm, which significantly improves efficiency and convergence speed through enhanced pheromone diffusion mechanisms and updating strategies, applied to global path planning. To adapt to dynamically changing environments and achieve more precise local path planning, an asymmetric strategy network TD3 algorithm (ATD3) is further proposed, which utilizes global path planning information within the strategy network only, creating a new hierarchical path planning algorithm—ADACO-ATD3. Simulation experiments demonstrate that the proposed algorithm significantly outperforms in terms of path length and number of iterations, effectively enhancing the mobile robot’s path planning performance in complex dynamic environments.

Large Language Models Meet Next-Generation Networking Technologies: A Review

The evolution of network technologies has significantly transformed global communication, information sharing, and connectivity. Traditional networks, relying on static configurations and manual interventions, face substantial challenges such as complex management, inefficiency, and susceptibility to human error. The rise of artificial intelligence (AI) has begun to address these issues by automating tasks like network configuration, traffic optimization, and security enhancements. Despite their potential, integrating AI models in network engineering encounters practical obstacles including complex configurations, heterogeneous infrastructure, unstructured data, and dynamic environments. Generative AI, particularly large language models (LLMs), represents a promising advancement in AI, with capabilities extending to natural language processing tasks like translation, summarization, and sentiment analysis. This paper aims to provide a comprehensive review exploring the transformative role of LLMs in modern network engineering. In particular, it addresses gaps in the existing literature by focusing on LLM applications in network design and planning, implementation, analytics, and management. It also discusses current research efforts, challenges, and future opportunities, aiming to provide a comprehensive guide for networking professionals and researchers. The main goal is to facilitate the adoption and advancement of AI and LLMs in networking, promoting more efficient, resilient, and intelligent network systems.

Experimental and numerical validation of high strain rate impact response and progressive damage of 3D orthogonal woven composites

Advanced three-dimensional (3D) woven composites for aerospace and automotive applications are commonly subjected to complex dynamic environments involving vibrations and impacts, resulting in examining their impact properties is extremely important. This paper first experimentally discussed the influences of strain rates, weft yarn densities and loading directions on the impact performances and failure mechanisms of 3D orthogonal woven composites (3DOWCs). Secondly, full-scale finite element models were developed to predict the stress distribution and interfacial damage evolution process. The predictions were well in agreement with the experimental results. This research revealed that the impact characteristics exhibited strain rate sensitivity. With increasing weft yarn densities, the high strain rate impact behaviors also improved. Particularly, the warp impact strength of 3DOWCs with a weft yarn density of 2 yarn/cm (W5-2) at 812 s−1 was 17.4% and 24.0% higher than that of 3DOWCs with a weft yarn density of 1.5 yarn/cm (W5-1) at 822 s−1. Meanwhile, warp impact strength consistently exceeded to that of the weft impact strength. Additionally, strain rates, weft yarn densities, and loading directions dramatically affected the stress distribution and interfacial damage evolution process of 3DOWCs. Significant warp yarns fracture and matrix cracking were the principal failure patterns in the warp impact, whereas the damage in the weft impact was dominated by localized fracture of weft yarns and interfacial debonding.

RSO-SLAM: A Robust Semantic Visual SLAM With Optical Flow in Complex Dynamic Environments

RSO-SLAM: A Robust Semantic Visual SLAM With Optical Flow in Complex Dynamic Environments

ASLAM-FD: a multi-sensor adaptive collaborative fusion SLAM framework based on degradation detection and deep reinforcement learning

Abstract Aiming at the difficulty in ensuring accuracy and performance of existing Simultaneous Localization And Mapping (SLAM) technology based on multi-sensor fusion in complex dynamic environments, a multi-sensor adaptive fusion SLAM framework based on degradation detection and deep reinforcement learning (ASLAM-FD) is proposed. This framework can achieve adaptive collaborative precise adjustment of fusion weights (FWs) based on the real-time self-degradation states and relative degradation states quantified by each sensor, and can adapt to different tightly coupled SLAM algorithms based on FWs. In addition, within this framework, the continuous quantification models for the degradation states of internal/external sensors with certain versatility (we refer to them as EX-DM and IN-DM) is proposed. These quantitative models can achieve continuous quantification of degradation states of various mainstream internal/external sensors. Based on the above sensor degradation state quantification models, this paper further proposes a deep reinforcement learning (DRL) network suitable for adaptive collaborative adjustment of FWs. This network focuses more on the temporal nature of sensor observation data and degradation states, making it more suitable for modeling relationships between data with temporal features. In the experimental section, we adapted the proposed ASLAM-FD to different multi-sensor fusion SLAM algorithms on multiple datasets, and compared it with multiple advanced fusion SLAM algorithms. We found that adapting ASLAM-FD can effectively improve the accuracy and performance of fusion SLAM in complex dynamic environments.

A Novel Hypersonic Target Trajectory Estimation Method Based on Long Short-Term Memory and a Multi-Head Attention Mechanism.

To address the complex maneuvering characteristics of hypersonic targets in adjacent space, this paper proposes an LSTM trajectory estimation method combined with the attention mechanism and optimizes the model from the information-theoretic perspective. The method captures the target dynamics by using the temporal processing capability of LSTM, and at the same time improves the efficiency of information utilization through the attention mechanism to achieve accurate prediction. First, a target dynamics model is constructed to clarify the motion behavior parameters. Subsequently, an LSTM model incorporating the attention mechanism is designed, which enables the model to automatically focus on key information fragments in the historical trajectory. In model training, information redundancy is reduced, and information validity is improved through feature selection and data preprocessing. Eventually, the model achieves accurate prediction of hypersonic target trajectories with limited computational resources. The experimental results show that the method performs well in complex dynamic environments with improved prediction accuracy and robustness, reflecting the potential of information theory principles in optimizing the trajectory prediction model.

Improved Multi-Sensor Fusion Dynamic Odometry Based on Neural Networks

High-precision simultaneous localization and mapping (SLAM) in dynamic real-world environments plays a crucial role in autonomous robot navigation, self-driving cars, and drone control. To address this dynamic localization issue, in this paper, a dynamic odometry method is proposed based on FAST-LIVO, a fast LiDAR (light detection and ranging)–inertial–visual odometry system, integrating neural networks with laser, camera, and inertial measurement unit modalities. The method first constructs visual–inertial and LiDAR–inertial odometry subsystems. Then, a lightweight neural network is used to remove dynamic elements from the visual part, and dynamic clustering is applied to the LiDAR part to eliminate dynamic environments, ensuring the reliability of the remaining environmental data. Validation of the datasets shows that the proposed multi-sensor fusion dynamic odometry can achieve high-precision pose estimation in complex dynamic environments with high continuity, reliability, and dynamic robustness.

YLS-SLAM: a real-time dynamic visual SLAM based on semantic segmentation

Purpose Traditional visual simultaneous localization and mapping (SLAM) systems are primarily based on the assumption that the environment is static, which makes them struggle with the interference caused by dynamic objects in complex industrial production environments. This paper aims to improve the stability of visual SLAM in complex dynamic environments through semantic segmentation and its optimization. Design/methodology/approach This paper proposes a real-time visual SLAM system for complex dynamic environments based on YOLOv5s semantic segmentation, named YLS-SLAM. The system combines semantic segmentation results and the boundary semantic enhancement algorithm. By recognizing and completing the semantic masks of dynamic objects from coarse to fine, it effectively eliminates the interference of dynamic feature points on the pose estimation and enhances the retention and extraction of prominent features in the background, thereby achieving stable operation of the system in complex dynamic environments. Findings Experiments on the Technische Universität München and Bonn data sets show that, under monocular and Red, Green, Blue - Depth modes, the localization accuracy of YLS-SLAM is significantly better than existing advanced dynamic SLAM methods, effectively improving the robustness of visual SLAM. Additionally, the authors also conducted tests using a monocular camera in a real industrial production environment, successfully validating its effectiveness and application potential in complex dynamic environment. Originality/value This paper combines semantic segmentation algorithms with boundary semantic enhancement algorithms to effectively achieve precise removal of dynamic objects and their edges, while ensuring the system's real-time performance, offering significant application value.

Modeling bio-inspired visual neural for detecting visual features of small- and wide-field moving targets synchronously from complex dynamic environments

Modeling bio-inspired visual neural for detecting visual features of small- and wide-field moving targets synchronously from complex dynamic environments

Preparation and characterization of mussel-inspired chitosan/polydopamine films and their feasibility for oral mucosa application

Oral mucosal lesions (OML), which represent a major public health issue worldwide, include any pathological changes in the oral mucosa, such as ulcers, pigmentation, and swelling. Due to its humid and dynamic complex environment, designing oral mucosal preparations poses significant challenges. Drawing inspiration from mussels, this study employed an eco-friendly one-pot strategy for the preparation of chitosan/polydopamine (CS/PDA) films. We demonstrated that CS-induced polymerization of dopamine monomers under acidic conditions, which might be attributed to the large number of hydrogen bonding sites of CS chains. PDA markedly enhances properties of the CS film and exhibits concentration dependence. At the concentration of 1 wt% PDA, the lap-shear strength and tensile strength of CS/PDA films reached 5.01 ± 0.24 kpa and 4.20 ± 0.78 kpa, respectively, indicating that the mucosal adhesion ability was significantly improved. In comparison with the single CS film, the swelling rate of CS/PDA film decreased by about 30 %. Rheological results also showed that the storage modulus returned to 93 % after cyclic large strain, while the single CS film only recovered to 73 %. Moreover, these films demonstrated good biocompatibility and enhanced oral ulcer healing in rats, providing a new and practical option for the local treatment of OML.

Family of 2:1 resonant quasi-periodic distant retrograde orbits in cislunar space

Given the current enthusiasm for lunar exploration, the 2:1 resonant distant retrograde orbit (DRO) in Earth-Moon space is of interest. To gain an in-depth understanding of the complex dynamic environment in cislunar space and provide more options for parking orbits, this paper investigates the existence of quasi-periodic orbits near the 2:1 resonant DRO in the circular restricted three-body problem (CR3BP). Firstly, the numerical computation approach, continuation strategy, and stability analysis method of quasi-periodic orbits are introduced. Then, addressing the primary challenges in the continuation progress, we have developed an adaptive continuation algorithm with automatic adjustment of the step size and the number of discrete points that characterize the invariant torus. Finally, two types of 2D quasi-DROs and their linear stability properties are explored. Using Poincaré sections, we investigated the boundaries of the maximum extent attainable by both 2D quasi-DRO families in the CR3BP at a specific Jacobi energy, confirming that both types of quasi-periodic families have reached their respective boundaries. The algorithm described in this paper is beneficial for facilitating the computation of quasi-periodic families and aids in discovering additional potential dynamical structures.

The Research on Intelligent News Advertisement Recommendation Algorithm Based on Prompt Learning in End-to-End Large Language Model Architecture

With the explosive growth of information on the internet, users are increasingly facing the problem of information overload, making precise news and ad recommendations an important area of research. While traditional recommendation algorithms can meet user needs to some extent, they still have limitations in dealing with complex and changing user behaviors and dynamic content environments. This paper addresses the shortcomings of existing news and ad recommendation systems by proposing an intelligent recommendation algorithm based on an end-to-end large language model architecture. Firstly, we utilize the BERT model as the foundation, leveraging its powerful text representation capabilities to achieve deep semantic understanding of news and ad content, thereby capturing more detailed content features. Secondly, we apply prompt learning to fine-tune the BERT model, designing specific prompts for the model to better understand the implicit needs and preferences of users. Finally, we integrate these steps into an end-to-end architecture, enabling the model to achieve automated learning and optimization throughout the entire process from input to output, thus improving the precision and efficiency of recommendations. Experimental results demonstrate that the proposed method significantly outperforms traditional methods in the task of news and ad recommendation, not only enhancing the accuracy and relevance of recommendations but also effectively improving the model's interpretability and flexibility. This research explores new possibilities for the application of large language models in recommendation systems.

Brain-like border ownership signals support prediction of natural videos.

To make sense of visual scenes, the brain must segment foreground from background. This is thought to be facilitated by neurons in the primate visual system that encode border ownership (BOS), i.e. whether a local border is part of an object on one or the other side of the border. It is unclear how these signals emerge in neural networks without a teaching signal of what is foreground and background. In this study, we investigated whether BOS signals exist in PredNet, a self-supervised artificial neural network trained to predict the next image frame of natural video sequences. We found that a significant number of units in PredNet are selective for BOS. Moreover these units share several other properties with the BOS neurons in the brain, including robustness to scene variations that constitute common object transformations in natural videos, and hysteresis of BOS signals. Finally, we performed ablation experiments and found that BOS units contribute more to prediction than non-BOS units for videos with moving objects. Our findings indicate that BOS units are especially useful to predict future input in natural videos, even when networks are not required to segment foreground from background. This suggests that BOS neurons in the brain might be the result of evolutionary or developmental pressure to predict future input in natural, complex dynamic visual environments.

Analysis of a Halocline Mesoscale Eddy in the Northwind Basin, Arctic Ocean

Eddies have been observed at all depths and in all regions of the Arctic Ocean. However, given the complex geographic conditions and dynamic environments of this ocean, the synchronized observations of temperature, salinity, and currents, and the detailed analysis of individual eddies are still lacking in the Northwind Basin. Our study aims to address these research gaps. We observed an eddy from a mooring in the Northwind Basin in late October 2017. It is a large anticyclonic cold eddy within the Arctic halocline, with a maximum azimuthal velocity reaching 52.63 cm/s and a horizontal scale (∼56 km) that significantly exceeds the first local baroclinic Rossby radius of deformation, i.e., it is in mesoscale. Its azimuthal velocity and scale are larger compared with those of nearby eddies, suggesting a relatively young state. This eddy possibly originated from the northern Chukchi Sea shelf, converging near Hanna Shoal with the Chukchi Slope Current before being advected northward into the Northwind Basin. Our study outlines detailed steps for extracting and analyzing eddies from mooring data and contributes to improving the understanding of the characteristics of Arctic Ocean eddies, providing a typical case for the investigation of eddies in the Northwind Basin.

MODELS OF REMOTE IDENTIFICATION OF PARAMETERS OF DYNAMIC OBJECTS USING DETECTION TRANSFORMERS AND OPTICAL FLOW

The tasks of remote identification of parameters of dynamic objects are important for various fields, including computer vision, robotics, autonomous vehicles, video surveillance systems, and many others. Traditional methods of solving these problems face the problems of insufficient accuracy and efficiency of determining dynamic parameters in conditions of rapidly changing environments and complex dynamic scenarios. Modern methods of identifying parameters of dynamic objects using technologies of detection transformers and optical flow are considered. Transformer detection is one of the newest approaches in computer vision that uses transformer architecture for object detection tasks. This transformer integrates the object detection and boundary detection processes into a single end-to-end model, which greatly improves the accuracy and speed of processing. The use of transformers allows the model to effectively process information from the entire image at the same time, which contributes to better recognition of objects even in difficult conditions. Optical flow is a motion analysis method that determines the speed and direction of pixel movement between successive video frames. This method allows obtaining detailed information about the dynamics of the scene, which is critical for accurate tracking and identification of parameters of moving objects. The integration of detection transformers and optical flow is proposed to increase the accuracy of identification of parameters of dynamic objects. The combination of these two methods allows you to use the advantages of both approaches: high accuracy of object detection and detailed information about their movement. The conducted experiments show that the proposed model significantly outperforms traditional methods both in the accuracy of determining the parameters of objects and in the speed of data processing. The key results of the study indicate that the integration of detection transformers and optical flow provides reliable and fast determination of parameters of moving objects in real time, which can be applied in various practical scenarios. The conducted research also showed the potential for further improvement of data processing methods and their application in complex dynamic environments. The obtained results open new perspectives for the development of intelligent monitoring and control systems capable of adapting to rapidly changing environmental conditions, increasing the efficiency and safety of their work.