Autonomous Intelligent Systems

Abstract The fault diagnosis of cargo UAVs (Unmanned Aerial Vehicles) is crucial to ensure the safety of logistics distribution. In the context of smart logistics, the new trend of utilizing knowledge graph (KG) for fault diagnosis is gradually emerging, bringing new opportunities to improve the efficiency and accuracy of fault diagnosis in the era of Industry 4.0. The operating environment of cargo UAVs is complex, and their faults are typically closely related to it. However, the available data only considers faults and maintenance data, making it difficult to diagnose faults accurately. Moreover, the existing KG suffers from the problem of confusing entity boundaries during the extraction process, which leads to lower extraction efficiency. Therefore, a fault diagnosis knowledge graph (FDKG) for cargo UAVs constructed based on multi-domain fusion and incorporating an attention mechanism is proposed. Firstly, the multi-domain ontology modeling is realized based on the multi-domain fault diagnosis concept analysis expression model and multi-dimensional similarity calculation method for cargo UAVs. Secondly, a multi-head attention mechanism is added to the BERT-BILSTM-CRF network model for entity extraction, relationship extraction is performed through ERNIE, and the extracted triples are stored in the Neo4j graph database. Finally, the DJI cargo UAV failure is taken as an example for validation, and the results show that the new model based on multi-domain fusion data is better than the traditional model, and the precision rate, recall rate, and F1 value can reach 87.52%, 90.47%, and 88.97%, respectively.

Ao Xiao

Wei Yan

Xumei Zhang

Ying Liu

Hua Zhang

Qi Liu

Multi-domain fusion for cargo UAV fault diagnosis knowledge graph construction

多域融合用于货运无人机故障诊断知识图谱构建

Abstract Textile dyeing requires optimizing combinations of ingredients and process parameters to achieve target colour properties. Modelling the complex relationships between these factors and the resulting colour is challenging. In this case, a physics-informed approach for multi-output regression to model CIELAB colour values from dyeing ingredient and process inputs is proposed. Leveraging attention mechanisms and multi-task learning, the model outperforms baseline methods at predicting multiple colour outputs jointly. Specifically, the Transformer model’s attention mechanism captures the complex interactions between dyeing ingredients and process parameters, while the multi-task learning framework exploits the intrinsic correlations among the L*, a*, and b* dimensions of the CIELAB colour space. In addition, the incorporation of physical knowledge through a physics-informed loss function integrates the CMC colour difference formula. This loss function, along with the attention mechanisms, enables the model to learn the nuanced relationships between the dyeing process variables and the final colour output, thereby improving the overall prediction accuracy. This reduces trial-and-error costs and resource waste, contributing to environmental sustainability by minimizing water and energy consumption and chemical emissions.

Zheyuan Chen

Jian Liu

Jian Li

Mukun Yuan

Guangping Yu

Leveraging multi-output modelling for CIELAB using colour difference formula towards sustainable textile dyeing

利用多输出建模与CIELAB色差公式实现可持续纺织染色

Abstract As the core competitiveness of the national industry, large-scale equipment such as ships, high-speed rail and nuclear power equipment, their production process involves in-depth personalization. It includes complex processes and long manufacturing cycles. In addition, the equipment’s supply chain management is extremely complex. Therefore, the development of a supply chain management knowledge graph is of significant strategic significance. It not only enhances the synergistic effect of the supply chain management but also upgrades the level of intelligent management. This paper proposes a distant supervision knowledge extraction and knowledge graph construction method in the supply chain management of large equipment manufacturing, which achieves digital and structured management and efficient use of supply chain management knowledge in the industry. This paper presents an approach to extract entity-relation knowledge using limited samples. We achieve this by establishing a distant supervision model. Furthermore, we introduce a fusion gate mechanism and integrate ontology information, thereby enhancing the model’s capability to effectively discern sentence-level semantics. Subsequently, we promptly modify the weights of input features using the gate mechanism to strengthen the model’s resilience and address the issue of vector noise diffusion. Finally, an inter-bag sentence attention mechanism is introduced to integrate different sentence bag information at the sentence bag level, which achieves more accurate entity-relation knowledge extraction. The experimental results prove that compared with the latest distant supervision method, the accuracy of relation extraction is improved by 2.8%, and the AUC value is increased by 3.9%, effectively improving the quality of knowledge graph in supply chain management.

Feiyue Huang

Lianglun Cheng

Distant supervision knowledge extraction and knowledge graph construction method for supply chain management domain

供应链管理领域远监督知识抽取与知识图谱构建方法

Abstract Machining quality prediction based on cutting big data is the core focus of current developments in intelligent manufacturing. Presently, predictions of machining quality primarily rely on process and signal analyses. Process-based predictions are generally constrained to the development of rudimentary regression models. Signal-based predictions often require large amounts of data, multiple processing steps (such as noise reduction, principal component analysis, modulation, etc.), and have low prediction efficiency. In addition, the accuracy of the model depends on tedious manual parameter tuning. This paper proposes a convolutional neural network quality intelligent prediction model based on automatic feature extraction and adaptive data fusion (CNN-AFEADF). Firstly, by processing signals from multiple directions, time-frequency domain images with rich features can be obtained, which significantly benefit neural network learning. Secondly, the corresponding images in three directions are fused into one image by setting different fusion weight parameters. The optimal fusion weight parameters and window length are determined by the Particle Swarm Optimization algorithm (PSO). This data fusion method reduces training time by 16.74 times. Finally, the proposed method is verified by various experiments. This method can automatically identify sensitive data features through neural network fitting experiments and optimization, thereby eliminating the need for expert experience in determining the significance of data features. Based on this approach, the model achieves an average relative error of 2.95%, reducing the prediction error compared to traditional models. Furthermore, this method enhances the intelligent machining level.

Xun Zhang

Sibao Wang

Fangrui Gao

Hao Wang

Haoyu Wu

An intelligent surface roughness prediction method based on automatic feature extraction and adaptive data fusion

基于自动特征提取与自适应数据融合的智能表面粗糙度预测方法

Abstract This paper introduces a motion planning and cooperative formation control approach for quadruped robots and multi-agent systems. First, in order to improve the efficiency and safety of quadruped robots navigating in complex environments, this paper proposes a new planning method that combines the dynamic model of quadruped robots and a gradient-optimized obstacle avoidance strategy without Euclidean Signed Distance Field. The framework is suitable for both static and slow dynamic obstacle environments, aiming to achieve multiple goals of obstacle avoidance, minimizing energy consumption, reducing impact, satisfying dynamic constraints, and ensuring trajectory smoothness. This approach differs in that it reduces energy consumption throughout the movement from a new perspective. Meanwhile, this method effectively reduces the impact of the ground on the robot, thus mitigating the damage to its structure. Second, we combine the dynamic control barrier function and the virtual leader-follower model to achieve efficient and safe formation control through model predictive control. Finally, the proposed algorithm is validated through both simulations and real-world scenarios testing.

Zongrui Ji

Yi Dong

Safe motion planning and formation control of quadruped robots

四足机器人安全运动规划与编队控制

Abstract Previous studies have primarily focused on converting point clouds (PC) into a dense mech of 3D finite element models, neglecting the conversion of PCs into as-built wireframe models with two-node elements for line elements such as beams and columns. This study aims to demonstrate the feasibility of this direct conversion, utilizing building framing patterns to create wireframe models. The study also integrates the OpenSeesPy package for modal analysis and double integration for bending estimation to demonstrate the application of the presented method in robotic inspection. Results indicate the successful conversion of a 4-story mass timber building PC to a 3D structural model with an average error of 7.5% under simplified assumptions. Further, two complex mass timber shed PCs were tested, resulting in detailed wireframe models. According to resource monitoring, our method can process ∼593 points/second, mostly affected by the number of neighbors used in the first stage of sparse points removal. Lastly, our method detects beams, columns, ceilings (floors), and walls with their directions. This research can facilitate various structural modeling directly based on PC data for digital twinning and autonomous robotic inspection.

Farzad Azizi Zade

Arvin Ebrahimkhanlou

Point clouds to as-built two-node wireframe digital twin: a novel method to support autonomous robotic inspection

点云到双节点线框数字孪生：一种支持自主机器人检测的新方法

Abstract Combining safety objectives with stability objectives is crucial for safety-critical systems. Existing studies generally unified these two objectives by constructing Lyapunov-type barrier functions. However, insufficient analysis of key set relationships within the system may render the proposed safety and stability conditions conservative, and these studies also did not provide how to use such conditions to design safety-stability control strategies. This paper proposed a feasible and constructive design to achieve stabilization of safety-critical systems by a relaxed converse Lyapunov-barrier approach. By analyzing the relationships between a series of sets associated with the safety-critical system, the stability and safety conditions can be appropriately relaxed. Then, with the help of relaxed converse control Lyapunov-barrier functions (RCCLBFs), a theoretical result was obtained for the stability of affine nonlinear systems with safety constraints. Subsequently, a constructive method was developed for a second-order strict-feedback system to transform the process of solving RCCLBFs into a Lyapunov-like stabilization problem. Finally, the proposed safety-stability control method is exerted on a robotic system and demonstrated by simulations.

Haoqi Li

Jiangping Hu

Xiaoming Hu

Bijoy K. Ghosh

Stabilization of nonlinear safety-critical systems by relaxed converse Lyapunov-barrier approach and its applications in robotic systems

通过松弛逆李雅普诺夫-障碍方法稳定非线性安全关键系统及其在机器人系统中的应用

Abstract As an essential part of modern smart manufacturing, road transport with large and heavy trucks has in-creased dramatically. Due to the inside wheel difference in the process of turning, there is a considerable safety hazard in the blind area of the inside wheel difference. In this paper, multiple cameras combined with deep learning algorithms are introduced to detect pedestrians in the blind area of wheel error. A scheme of vehicle-pedestrian safety alarm detection system is developed via the integration of YOLOv5 and an improved binocular distance measurement method. The system accurately measures the distance between the truck and nearby pedestrians by utilizing multiple cameras and PP Human recognition, providing real-time safety alerts. The experimental results show that this method significantly reduces distance measurement errors, improves the reliability of pedestrian detection, achieves high accuracy and real-time performance, and thus enhances the safety of trucks in complex traffic environments.

Tingting Bao

Ding Lin

Zhiguo Zhou

Kejia Wang

Pedestrian safety alarm system based on binocular distance measurement for trucks using recognition feature analysis

基于双目测距的卡车行人安全报警系统：识别特征分析

Publication Information	Publisher: Springer，Publishing cycle: ，Journal Type: journal，Open Access Journals: Yes
Basic data	Year of publication: 2021，Proportion of original research papers: ，Self Citation Rate:， Gold OA Rate:
Average review cycle	21 weeks

Autonomous Intelligent Systems

Journal Citation Format

A journal article with 1 author

A journal article with 2 authors

A journal article with 3 authors

A journal article with 5 or more authors

Books Citation Format

Thesis Citation Format

Web sites Citation Format

Patent Citation Format

Staying up late manually editing references? ivySCI automatically matches journals and helps you generate references with a single click.

Share Submission Experience

Multi-domain fusion for cargo UAV fault diagnosis knowledge graph construction

Leveraging multi-output modelling for CIELAB using colour difference formula towards sustainable textile dyeing

Distant supervision knowledge extraction and knowledge graph construction method for supply chain management domain

An intelligent surface roughness prediction method based on automatic feature extraction and adaptive data fusion

Safe motion planning and formation control of quadruped robots

An intelligent surface roughness prediction method based on automatic feature extraction and adaptive data fusion

Safe motion planning and formation control of quadruped robots

Point clouds to as-built two-node wireframe digital twin: a novel method to support autonomous robotic inspection

Stabilization of nonlinear safety-critical systems by relaxed converse Lyapunov-barrier approach and its applications in robotic systems

Pedestrian safety alarm system based on binocular distance measurement for trucks using recognition feature analysis

ISSN	2730-616X
h-index	8