CAAI Transactions on Intelligence Technology最新文献

{"title":"Longitudinal velocity control of autonomous driving based on extended state observer","authors":"Hongbo Gao,&nbsp;Hanqing Yang,&nbsp;Xiaoyu Zhang,&nbsp;Xiangyun Ren,&nbsp;Fenghua Liang,&nbsp;Ruidong Yan,&nbsp;Qingchao Liu,&nbsp;Mingmao Hu,&nbsp;Fang Zhang,&nbsp;Jiabing Gao,&nbsp;Siyu Bao,&nbsp;Keqiang Li,&nbsp;Deyi Li,&nbsp;Danwei Wang","doi":"10.1049/cit2.12397","DOIUrl":"https://doi.org/10.1049/cit2.12397","url":null,"abstract":"<p>Active Disturbance Rejection Control (ADRC) possesses robust disturbance rejection capabilities, making it well-suited for longitudinal velocity control. However, the conventional Extended State Observer (ESO) in ADRC fails to fully exploit feedback from first-order and higher-order estimation errors and tracking error simultaneously, thereby diminishing the control performance of ADRC. To address this limitation, an enhanced car-following algorithm utilising ADRC is proposed, which integrates the improved ESO with a feedback controller. In comparison to the conventional ESO, the enhanced version effectively utilises multi-order estimation and tracking errors. Specifically, it enhances convergence rates by incorporating feedback from higher-order estimation errors and ensures the estimated value converges to the reference value by utilising tracking error feedback. The improved ESO significantly enhances the disturbance rejection performance of ADRC. Finally, the effectiveness of the proposed algorithm is validated through the Lyapunov approach and experiments.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"36-46"},"PeriodicalIF":8.4,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12397","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143533419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"KitWaSor: Pioneering pre-trained model for kitchen waste sorting with an innovative million-level benchmark dataset","authors":"Leyuan Fang,&nbsp;Shuaiyu Ding,&nbsp;Hao Feng,&nbsp;Junwu Yu,&nbsp;Lin Tang,&nbsp;Pedram Ghamisi","doi":"10.1049/cit2.12399","DOIUrl":"https://doi.org/10.1049/cit2.12399","url":null,"abstract":"<p>Intelligent sorting is an important prerequisite for the full quantitative consumption and harmless disposal of kitchen waste. The existing object detection method based on an ImageNet pre-trained model is an effective way of sorting. Owing to significant domain gaps between natural images and kitchen waste images, it is difficult to reflect the characteristics of diverse scales and dense distribution in kitchen waste based on an ImageNet pre-trained model, leading to poor generalisation. In this article, the authors propose the first pre-trained model for kitchen waste sorting called KitWaSor, which combines both contrastive learning (CL) and masked image modelling (MIM) through self-supervised learning (SSL). First, to address the issue of diverse scales, the authors propose a mixed masking strategy by introducing an incomplete masking branch based on the original random masking branch. It prevents the complete loss of small-scale objects while avoiding excessive leakage of large-scale object pixels. Second, to address the issue of dense distribution, the authors introduce semantic consistency constraints on the basis of the mixed masking strategy. That is, object semantic reasoning is performed through semantic consistency constraints to compensate for the lack of contextual information. To train KitWaSor, the authors construct the first million-level kitchen waste dataset across seasonal and regional distributions, named KWD-Million. Extensive experiments show that KitWaSor achieves state-of-the-art (SOTA) performance on the two most relevant downstream tasks for kitchen waste sorting (i.e. image classification and object detection), demonstrating the effectiveness of the proposed KitWaSor.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"94-114"},"PeriodicalIF":8.4,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12399","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143536116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-station multi-robot task assignment method based on deep reinforcement learning","authors":"Junnan Zhang,&nbsp;Ke Wang,&nbsp;Chaoxu Mu","doi":"10.1049/cit2.12394","DOIUrl":"https://doi.org/10.1049/cit2.12394","url":null,"abstract":"<p>This paper focuses on the problem of multi-station multi-robot spot welding task assignment, and proposes a deep reinforcement learning (DRL) framework, which is made up of a public graph attention network and independent policy networks. The graph of welding spots distribution is encoded using the graph attention network. Independent policy networks with attention mechanism as a decoder can handle the encoded graph and decide to assign robots to different tasks. The policy network is used to convert the large scale welding spots allocation problem to multiple small scale single-robot welding path planning problems, and the path planning problem is quickly solved through existing methods. Then, the model is trained through reinforcement learning. In addition, the task balancing method is used to allocate tasks to multiple stations. The proposed algorithm is compared with classical algorithms, and the results show that the algorithm based on DRL can produce higher quality solutions.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"134-146"},"PeriodicalIF":8.4,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12394","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143533414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-omics graph convolutional networks for digestive system tumour classification and early-late stage diagnosis","authors":"Lin Zhou,&nbsp;Zhengzhi Zhu,&nbsp;Hongbo Gao,&nbsp;Chunyu Wang,&nbsp;Muhammad Attique Khan,&nbsp;Mati Ullah,&nbsp;Siffat Ullah Khan","doi":"10.1049/cit2.12395","DOIUrl":"https://doi.org/10.1049/cit2.12395","url":null,"abstract":"<p>The prevalence of digestive system tumours (DST) poses a significant challenge in the global crusade against cancer. These neoplasms constitute 20% of all documented cancer diagnoses and contribute to 22.5% of cancer-related fatalities. The accurate diagnosis of DST is paramount for vigilant patient monitoring and the judicious selection of optimal treatments. Addressing this challenge, the authors introduce a novel methodology, denominated as the Multi-omics Graph Transformer Convolutional Network (MGTCN). This innovative approach aims to discern various DST tumour types and proficiently discern between early-late stage tumours, ensuring a high degree of accuracy. The MGTCN model incorporates the Graph Transformer Layer framework to meticulously transform the multi-omics adjacency matrix, thereby illuminating potential associations among diverse samples. A rigorous experimental evaluation was undertaken on the DST dataset from The Cancer Genome Atlas to scrutinise the efficacy of the MGTCN model. The outcomes unequivocally underscore the efficiency and precision of MGTCN in diagnosing diverse DST tumour types and successfully discriminating between early-late stage DST cases. The source code for this groundbreaking study is readily accessible for download at https://github.com/bigone1/MGTCN.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"9 6","pages":"1572-1586"},"PeriodicalIF":8.4,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12395","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143247961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving long-tail classification via decoupling and regularisation","authors":"Shuzheng Gao,&nbsp;Chaozheng Wang,&nbsp;Cuiyun Gao,&nbsp;Wenjian Luo,&nbsp;Peiyi Han,&nbsp;Qing Liao,&nbsp;Guandong Xu","doi":"10.1049/cit2.12374","DOIUrl":"https://doi.org/10.1049/cit2.12374","url":null,"abstract":"<p>Real-world data always exhibit an imbalanced and long-tailed distribution, which leads to poor performance for neural network-based classification. Existing methods mainly tackle this problem by reweighting the loss function or rebalancing the classifier. However, one crucial aspect overlooked by previous research studies is the imbalanced feature space problem caused by the imbalanced angle distribution. In this paper, the authors shed light on the significance of the angle distribution in achieving a balanced feature space, which is essential for improving model performance under long-tailed distributions. Nevertheless, it is challenging to effectively balance both the classifier norms and angle distribution due to problems such as the low feature norm. To tackle these challenges, the authors first thoroughly analyse the classifier and feature space by decoupling the classification logits into three key components: classifier norm (i.e. the magnitude of the classifier vector), feature norm (i.e. the magnitude of the feature vector), and cosine similarity between the classifier vector and feature vector. In this way, the authors analyse the change of each component in the training process and reveal three critical problems that should be solved, that is, the imbalanced angle distribution, the lack of feature discrimination, and the low feature norm. Drawing from this analysis, the authors propose a novel loss function that incorporates hyperspherical uniformity, additive angular margin, and feature norm regularisation. Each component of the loss function addresses a specific problem and synergistically contributes to achieving a balanced classifier and feature space. The authors conduct extensive experiments on three popular benchmark datasets including CIFAR-10/100-LT, ImageNet-LT, and iNaturalist 2018. The experimental results demonstrate that the authors’ loss function outperforms several previous state-of-the-art methods in addressing the challenges posed by imbalanced and long-tailed datasets, that is, by improving upon the best-performing baselines on CIFAR-100-LT by 1.34, 1.41, 1.41 and 1.33, respectively.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"62-71"},"PeriodicalIF":8.4,"publicationDate":"2024-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12374","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143536070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-sensor missile-borne LiDAR point cloud data augmentation based on Monte Carlo distortion simulation","authors":"Luda Zhao,&nbsp;Yihua Hu,&nbsp;Fei Han,&nbsp;Zhenglei Dou,&nbsp;Shanshan Li,&nbsp;Yan Zhang,&nbsp;Qilong Wu","doi":"10.1049/cit2.12389","DOIUrl":"https://doi.org/10.1049/cit2.12389","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <p>Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks. Due to the diversity and robustness constraints of the data, data augmentation (DA) methods are utilised to expand dataset diversity and scale. However, due to the complex and distinct characteristics of LiDAR point cloud data from different platforms (such as missile-borne and vehicular LiDAR data), directly applying traditional 2D visual domain DA methods to 3D data can lead to networks trained using this approach not robustly achieving the corresponding tasks. To address this issue, the present study explores DA for missile-borne LiDAR point cloud using a Monte Carlo (MC) simulation method that closely resembles practical application. Firstly, the model of multi-sensor imaging system is established, taking into account the joint errors arising from the platform itself and the relative motion during the imaging process. A distortion simulation method based on MC simulation for augmenting missile-borne LiDAR point cloud data is proposed, underpinned by an analysis of combined errors between different modal sensors, achieving high-quality augmentation of point cloud data. The effectiveness of the proposed method in addressing imaging system errors and distortion simulation is validated using the imaging scene dataset constructed in this paper. Comparative experiments between the proposed point cloud DA algorithm and the current state-of-the-art algorithms in point cloud detection and single object tracking tasks demonstrate that the proposed method can improve the network performance obtained from unaugmented datasets by over 17.3% and 17.9%, surpassing SOTA performance of current point cloud DA algorithms.</p>\u0000 </section>\u0000 </div>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"300-316"},"PeriodicalIF":8.4,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12389","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143533353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Resource-adaptive and OOD-robust inference of deep neural networks on IoT devices","authors":"Cailen Robertson,&nbsp;Ngoc Anh Tong,&nbsp;Thanh Toan Nguyen,&nbsp;Quoc Viet Hung Nguyen,&nbsp;Jun Jo","doi":"10.1049/cit2.12384","DOIUrl":"https://doi.org/10.1049/cit2.12384","url":null,"abstract":"<p>Efficiently executing inference tasks of deep neural networks on devices with limited resources poses a significant load in IoT systems. To alleviate the load, one innovative method is branching that adds extra layers with classification exits to a pre-trained model, enabling inputs with high-confidence predictions to exit early, thus reducing inference cost. However, branching networks, not originally tailored for IoT environments, are susceptible to noisy and out-of-distribution (OOD) data, and they demand additional training for optimal performance. The authors introduce BrevisNet, a novel branching methodology designed for creating on-device branching models that are both resource-adaptive and noise-robust for IoT applications. The method leverages the refined uncertainty estimation capabilities of Dirichlet distributions for classification predictions, combined with the superior OOD detection of energy-based models. The authors propose a unique training approach and thresholding technique that enhances the precision of branch predictions, offering robustness against noise and OOD inputs. The findings demonstrate that BrevisNet surpasses existing branching techniques in training efficiency, accuracy, overall performance, and robustness.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"115-133"},"PeriodicalIF":8.4,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12384","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143535724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A criterion for selecting the appropriate one from the trained models for model-based offline policy evaluation","authors":"Chongchong Li,&nbsp;Yue Wang,&nbsp;Zhi-Ming Ma,&nbsp;Yuting Liu","doi":"10.1049/cit2.12376","DOIUrl":"https://doi.org/10.1049/cit2.12376","url":null,"abstract":"<p>Offline policy evaluation, evaluating and selecting complex policies for decision-making by only using offline datasets is important in reinforcement learning. At present, the model-based offline policy evaluation (MBOPE) is widely welcomed because of its easy to implement and good performance. MBOPE directly approximates the unknown value of a given policy using the Monte Carlo method given the estimated transition and reward functions of the environment. Usually, multiple models are trained, and then one of them is selected to be used. However, a challenge remains in selecting an appropriate model from those trained for further use. The authors first analyse the upper bound of the difference between the approximated value and the unknown true value. Theoretical results show that this difference is related to the trajectories generated by the given policy on the learnt model and the prediction error of the transition and reward functions at these generated data points. Based on the theoretical results, a new criterion is proposed to tell which trained model is better suited for evaluating the given policy. At last, the effectiveness of the proposed criterion is demonstrated on both benchmark and synthetic offline datasets.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"223-234"},"PeriodicalIF":8.4,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12376","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143535725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pre-trained SAM as data augmentation for image segmentation","authors":"Junjun Wu,&nbsp;Yunbo Rao,&nbsp;Shaoning Zeng,&nbsp;Bob Zhang","doi":"10.1049/cit2.12381","DOIUrl":"https://doi.org/10.1049/cit2.12381","url":null,"abstract":"<p>Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset. Initially, data augmentation mainly involved some simple transformations of images. Later, in order to increase the diversity and complexity of data, more advanced methods appeared and evolved to sophisticated generative models. However, these methods required a mass of computation of training or searching. In this paper, a novel training-free method that utilises the Pre-Trained Segment Anything Model (SAM) model as a data augmentation tool (PTSAM-DA) is proposed to generate the augmented annotations for images. Without the need for training, it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved annotations. In this way, annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation model. Multiple comparative experiments on three datasets are conducted, including an in-house dataset, ADE20K and COCO2017. On this in-house dataset, namely Agricultural Plot Segmentation Dataset, maximum improvements of 3.77% and 8.92% are gained in two mainstream metrics, mIoU and mAcc, respectively. Consequently, large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"268-282"},"PeriodicalIF":8.4,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12381","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143536006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Grey-box modelling for estimation of optimum cut point temperature of crude distillation column","authors":"Junaid Shahzad,&nbsp;Iftikhar Ahmad,&nbsp;Muhammad Ahsan,&nbsp;Farooq Ahmad,&nbsp;Husnain Saghir,&nbsp;Manabu Kano,&nbsp;Hakan Caliskan,&nbsp;Hiki Hong","doi":"10.1049/cit2.12386","DOIUrl":"https://doi.org/10.1049/cit2.12386","url":null,"abstract":"<p>A grey-box modelling framework was developed for the estimation of cut point temperature of a crude distillation unit (CDU) under uncertainty in crude composition and process conditions. First principle (FP) model of CDU was developed for Pakistani crudes from Zamzama and Kunnar fields. A hybrid methodology based on the integration of Taguchi method and genetic algorithm (GA) was employed to estimate the optimal cut point temperature for various sets of process variables. Optimised datasets were utilised to develop an artificial neural networks (ANN) model for the prediction of optimum values of cut points. The ANN model was then used to replace the hybrid framework of the Taguchi method and the GA. The integration of the ANN and FP model makes it a grey-box (GB) model. For the case of Zamama crude, the GB model helped in the decrease of up to 38.93% in energy required per kilo barrel of diesel and an 8.2% increase in diesel production compared to the stand-alone FP model under uncertainty. Similarly, for Kunnar crude, up to 18.87% decrease in energy required per kilo barrel of diesel and a 33.96% increase in diesel production was observed in comparison to the stand-alone FP model.</p>","PeriodicalId":46211,"journal":{"name":"CAAI Transactions on Intelligence Technology","volume":"10 1","pages":"160-174"},"PeriodicalIF":8.4,"publicationDate":"2024-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/cit2.12386","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143533415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}