Shaofei Zang

Unsupervised Domain Adaptation (UDA), a key subfield of Transfer Learning (TL), addresses the performance degradation caused by domain shift that occurs when a model trained on a labeled source domain is applied to an unlabeled target domain. Despite considerable advances, most existing UDA methods suffer from two major limitations: (1) heavy reliance on extracting domain-invariant features without explicitly enhancing their discriminative power for classification or recognition tasks, and (2) confinement to a single adaptation strategy-such as feature-based, instance-based, or parameter-based-overlooking the potential advantages of combining multiple adaptation paradigms. To tackle these issues simultaneously, this paper proposes a novel Hybrid Adaptation of Feature, Parameter, and Instance (HA-FPI) framework for UDA. HA-FPI integrates feature alignment, parameter adaptation, and instance selection into a unified optimization objective, facilitating the extraction of cross-domain invariant yet discriminative features, optimizing shared parameters, and selecting high-confidence target samples. This integrated approach mitigates feature distribution discrepancy and significantly enhances cross-domain knowledge transfer efficiency. The proposed method comprises three stages: discriminative cross-domain invariant feature extraction, classifier parameter adaptation, and fused decision-making. In the first stage, the Cross-Domain Mean Approximation (CDMA) distance is minimized to identify a low-dimensional feature subspace that reduces both marginal and conditional distribution discrepancy between domains. Additionally, CDMA is combined with linear interpolation to construct a cross-domain linear sample generation mechanism, producing transitional samples that smooth the knowledge transfer process. To improve the discriminability of the shared feature representations, a discriminative penalty constraint is applied to both original and synthesized samples, effectively separating heterogeneous samples in the feature subspace. During the classifier parameter adaptation stage, a Least Squares Regression (LSR) is trained on source samples within the shared feature subspace. A unidirectional CDMA is incorporated into its objective function, yielding a source LSR with inherent transfer capability. Using CDMA as a confidence measure, high-confidence target samples are selected and combined with source samples to train a target LSR. In the final fused decision stage, predictions for the test samples are obtained through the joint decisions of both LSRs. Extensive experiments on benchmark domain adaptation datasets including Office+Caltech256, Office-31, USPS+MNIST, and ImageCLEF demonstrate the superior performance of the proposed HA-FPI framework compared to state-of-the-art UDA methods.

In this paper, an adaptive fixed/arbitrary‐time distributed cooperative guidance law is proposed to solve the three‐dimensional simultaneous attack problem. Firstly, with undirected topologies, a second‐order leader‐following consensus protocol, integral sliding mode, and adaptive law are implemented to construct an adaptive fixed‐time cooperative control law for intercepting a maneuvering target. Range‐to‐go is chosen as the consensus variable instead of time‐to‐go, and simultaneous arrival of multiple missiles is achieved for the incoming target despite the variations in velocity magnitude. Meanwhile, a robust fast fixed‐time observer is designed to estimate target accelerations, and an arbitrary‐time sliding mode manifold is constructed and maintained across the entire time domain, while a continuous adaptive sliding mode control law is formulated through synergistic integration of adaptation law and quasiswitching function. The settling time of the closed‐loop guidance system can be precisely chosen by a predetermined time constant, and LOS angular rates can be effectively steered to the equilibrium point. By relying on the position and velocity of the incoming target, normal acceleration and longitudinal acceleration can achieve a simultaneous hit‐to‐kill attack with the desired impact angles, while fast fixed/arbitrary‐time stability is proved through the Lyapunov theory. Finally, numerical simulations are performed to validate the feasibility and effectiveness of the theoretical results.

In the context of multi-destination material transportation tasks for unmanned aerial vehicles (UAVs) during wartime, task allocation and path planning play crucial roles. Although numerous research methods have been proposed, most algorithms treat these two aspects as relatively independent components, resulting in limited flexibility in decision-making and real-time path planning on highly confrontational battlefields. To address this issue, this paper presents a novel approach to multi-destination path planning based on ant colony decision-making and rolling control. Firstly, an information sharing module is established to process real-time updated environmental data, providing a foundation for decision-making and path planning. Secondly, a new ant colony decision method is developed by designing an objective function that minimizes consumption to determine the order of destination material transportation. Additionally, an online and multi-round destination decision strategy is formulated based on the updated environmental information from the information sharing module. This strategy prevents situations where total consumption for multiple destinations becomes excessive due to UAVs not considering real-time changes in material demand during transportation but making decisions only once per destination. Subsequently, the receding horizon control with extended solution method (RHC-eS) is employed to achieve multi-destination path planning by integrating online path planning and decision strategies. Finally, experiments are conducted in a multi-destination environment with obstacles 36 to validate the effectiveness of our proposed method. The results demonstrate that compared to several other advanced optimization methods, our approach exhibits advantages such as shorter paths and lower overall material consumption.

The problem of cooperative interception of the manoeuvring target is investigated in this paper. Firstly, in light of fast fixed-time consensus theory, time-to-go, and undirected topologies, adaptive cooperative guidance along the line-of-sight (LOS) direction is proposed to guarantee impact time synchronization. Next, novel nonsingular terminal sliding mode (NTSM) is designed to establish adaptive fixed-time guidance law for steering LOS angular rates to the origin or its small neighbourhood. Without the knowledge of target manoeuvre, the proposed cooperative guidance law can be provided by lateral and longitudinal accelerations of each missile, while more reasonable and rigorous analysis of fixed-time stability is carried out through the Lyapunov theory. Within the specified time, both control tasks of simultaneous attack and the desired impact angles can be completed before the final time of the guidance process. Finally, numerical simulations demonstrate the feasibility and effectiveness of the proposed scheme.

The recognition of oil-bearing formation is an important part in oil exploration, and recognition technology influences the predictive accuracy and efficiency. Low rank representation (LRR) has aroused much attention in the field of data mining. As a modified version, the low rank representation with adaptive graph regularization (LRR-AGR) exploits the global and local information of data for graph learning, and it simultaneously integrates distance regularization term, non-negative constraint and a rank constraint into the framework of LRR. However, how to balance these regularization terms according to the data greatly limits its clustering performance. To adaptively balance these regularization terms according to data and further improve the clustering performance, we propose a novel model named low-rank representation with adaptive parameters and graph regularization (LRR-APGR) in this paper. Firstly, a novel parameter optimization model is formulated and designed based on the framework of LRR-AGR and the feedback mechanism. Secondly, two global intelligent optimization algorithms, which can effectively solve the parameter optimization problem are presented based on particle swarm optimization (PSO) in multi-dimensional continuous space. Experimental results on the data oilsk81, oilsk83 and oilsk85 wells of Jianghan oil fields in China show that the proposed method can significantly improve the clustering performance and the predictive accuracy. DOI: 10.61416/ceai.v25i3.8650

With fast learning speed and high accuracy, extreme learning machine (ELM) has achieved great success in pattern recognition and machine learning. Unfortunately, it will fail in the circumstance where plenty of labeled samples for training model are insufficient. The labeled samples are difficult to obtain due to their high cost. In this paper, we solve this problem with transfer learning and propose joint transfer extreme learning machine (JTELM). First, it applies cross-domain mean approximation (CDMA) to minimize the discrepancy between domains, thus obtaining one ELM model. Second, subspace alignment (sa) and weight approximation are together introduced into the output layer to enhance the capability of knowledge transfer and learn another ELM model. Third, the prediction of test samples is dominated by the two learned ELM models. Finally, a series of experiments are carried out to investigate the performance of JTELM, and the results show that it achieves efficiently the task of transfer learning and performs better than the traditional ELM and other transfer or nontransfer learning methods.

For the guidance problem of multiple missiles cooperatively intercepting a maneuvering target, fixed-time cooperative guidance law is proposed with impact angle constraints. First, fixed-time consensus, integral sliding mode technique, and adaptive super-twisting algorithm are implemented to construct cooperative guidance law along the line-of-sight (LOS) direction. Second, based on fixed-time convergence observer and terminal sliding mode control, one fixed-time guidance law along the normal direction of LOS is designed to ensure that the missiles attack the target with expected angles. Finally, simulation results are carried out to demonstrate the effectiveness of the proposed fixed-time cooperative guidance law.