最新刊期
卷 7 ， 期 1 ， 2025
- 摘要：Undoubtedly, DeepSeek would speed up AI research, development, and applications. However, we need to think deeply and must intensify our effort in AI governance related investigation and implementation before important issues become out of control.关键词：AI governance;DeepSeek;DeepThink434|358|2更新时间：2025-03-15
Review Intelligence
- 摘要：In recent years, end-to-end solutions integrating perception and decision-making have made groundbreaking progress, providing new approaches to enhancing the safety and reliability of autonomous driving. Existing reviews often focus on the perception of the external environment by individual vehicles, overlooking the complex interactions in multi-agent collaboration. Based on the current state of research, integrated perception and decision-making methods for single agents were explored from an application perspective, and methods under multi-agent collaborative conditions were analyzed. Firstly, foundational technologies for integrated perception and decision-making in autonomous driving were summarized. Secondly, the latest research advancements in integrated perception and decision-making technologies for autonomous driving were highlighted. Thirdly, large-scale public datasets commonly used in the field of autonomous driving were elaborated upon. Then, the evaluation methods currently employed for integrated perception and decision-making in autonomous driving were introduced. Finally, a summary and outlook on integrated perception and decision-making in autonomous driving were presented.关键词：intelligent transportation;autonomous driving;integrated perception and decision-making1311|843|1更新时间：2025-03-15
- 摘要：New perspectives and powerful tools for criminology research were provided by exploring how the ACP (artificial systems, computational experiments, and parallel execution) approach and large language model (LLM) could be integrated into criminological studies. Firstly, an in-depth analysis of the challenges faced by traditional criminology in system characterization, dynamic simulation, and empirical research was conducted. The potential advantages of the ACP approach in addressing these challenges were then elaborated. Subsequently, a criminology research framework based on ACP and LLM was proposed, and its theoretical significance and practical value were discussed. Finally, it was envisioned how LLM-based digital assistants could enhance criminological research in the future. Given the challenges in practical applications, interdisciplinary collaboration and technological innovation are necessary to further advance criminological studies and improve crime prevention effectiveness.关键词：ACP approach;criminology;crime prevention;large language model249|393|0更新时间：2025-03-15
Surveys and Prospectives
- 摘要：The Chinese Association of Automation (CAA) is one of the most prominent scientific and technological association in the field of automation in China. Confronted with an increasingly complex scientific research environment and arduous research tasks, the CAA needs to constantly adjust its traditional organizational and operational models to meet new challenges. Based on smart contracts, decentralized autonomous organization and operation (DAO), and artificial intelligence technology represented by large models, the concept of "digital CAA" was proposed. The general architecture and key supporting technologies of digital CAA were presented. Its operational logic was described, its typical application scenarios were analyzed. Besides, the branch management was presented as a typical example to illustrate the auxiliary role digital CAA plays in helping CAA optimize their management and operation. Finally, the main challenges faced by digital CAA were analyzed. It aims to enhance the sustainability of CAA and serve as a reference model for similar transformations in other scientific and technological institutions.关键词：Chinese Association of Automation;decentralized autonomous organization and operation;scientific and technological association;smart contract318|676|0更新时间：2025-03-15
- 摘要：With the rapid development of road traffic infrastructure and vehicle connectivity, the construction of a safe, flexible, and efficient next-generation traffic control system for large-scale road networks has become a critical issue in the industry. To address the challenges of system connectivity, coordination, and scalability in current traffic control systems caused by heterogeneous infrastructure, tight coupling of software and hardware, and a predominantly physical control approach, this paper introduces a software-defined programmable traffic control system based on the concept of parallel transportation with virtual-physical interaction. This system integrated centralized controllers, an application development framework, traffic facility virtualization, and parallel control. These features enable efficient coordination and flexible scheduling of intelligent and connected traffic entities. First, the concept and characteristics of programmable traffic control systems were introduced. Then, the architecture and key technologies were discussed. Furthermore, the system's applications in traffic signal control, priority for emergency vehicles, adaptive traffic sign and marking control, and connected vehicle guidance were discussed. Finally, the development prospects of programmable traffic control systems and related technologies were summarized and prospected. This system can support the next-generation of traffic control systems and provide a foundation for the digital and intelligent transformation of traffic governance.关键词：programmable traffic control system;parallel transportation system;digital twin;intelligent transportation system;cyber-physical-social system288|311|0更新时间：2025-03-15
- 摘要：The graph neural networks mainly focus on the syntactic structure when they are used to model the syntactic dependency tree of a sentence for aspect-based sentiment analysis (ABSA). However, these methods neglect the study of dependency tree itself and semantic information, and syntactic information in unstructured text can bring additional noise. To address these issues, a graph convolutional network (GCN) model integrating syntactic and semantic enhancements was proposed. Attention mechanism was used to extract aspect information and global information, and enhance semantic information through gating mechanism. Simultaneously, dependency tree was reshaped to increase the weight of opinion words and enhance syntactic information by utilizing external emotional knowledge and the distance features of words. Finally, dynamic feature fusion was performed on semantic and syntactic information. Extensive experiments were conducted on three public datasets. The experiment results show that the accuracy and macro-F1 values are better than the compared models, which indicate the effectiveness of the fusion of syntactic enhancement and semantic enhancement for ABSA.关键词：aspect-based sentiment analysis;graph convolutional network;attention mechanism;gating mechanism361|263|0更新时间：2025-03-15
- 摘要：In response to the issue of biased comprehensive evaluation results caused by anomalous ratings in comprehensive evaluation problems, an intelligent comprehensive scoring approach based on improved differential evolution algorithm had been proposed, which enhanced the fairness of comprehensive evaluation results by eliminating anomalous ratings. Firstly, considering the causes of abnormal scores, the fairness index and objective function were defined for the comprehensive evaluation problem. The constraint and the criterion of abnormal scores were also analyzed. Secondly, an intelligent comprehensive scoring approach had been proposed based on improved differential evolution algorithm, which introduced the following three improvements to the traditional differential evolution algorithm framework: a real encoding and decoding method was developed based on the score weights. The weighting method was used to write the constraints of the score weights into the objective function. As a result, the comprehensive evaluation problem was transformed into an unconstrained optimization problem to improve efficiency. The problem-specific knowledge of score weights was designed based on the fairness index. A heuristic population initialization method was employed to speed up the algorithm convergence in the early stage and the solution quality of the score weights. Finally, the case study results had demonstrated that the proposed intelligent comprehensive scoring approach could automatically identify ratings with smaller weights as anomalous and eliminate them, ultimately providing a more fair comprehensive evaluation result within 13 seconds, effectively improving the fairness, impartiality, and scientific rationality of the comprehensive evaluation results.关键词：comprehensive evaluation;comprehensive scoring approach;differential evolution;evolutionary algorithm;expert review;abnormal score145|317|0更新时间：2025-03-15
Papers and Reports
- 摘要：Improving the positioning accuracy and real-time performance of disinfection robots are important prerequisites for achieving embodied intelligence, which is crucial to achieve a high sterilization rate of pathogenic microorganism, and can provide a strong guarantee for the health and safety of hospitals, funeral homes and other places. Therefore, a method for kinematics calibration and accuracy compensation of disinfection robot is proposed. First, a modeling method of kinematic model and error model based on screw theory is proposed. Second, the moving region is divided and the feature point data is sampled. Third, an error identification algorithm based on least square method is proposed, and the identification is realized by sampling data. Finally, the positioning error is compensated to the end position of the robot, and the accuracy compensation is realized. The correctness of the proposed method is verified by simulation experiments. The kinematic calibration and accuracy compensation method proposed can greatly improve the positioning accuracy of disinfection robot while ensuring real-time performance, enhance the level of robot's embodied intelligence, and provide a strong guarantee for improving the killing rate of germs and realizing reliable prevention of special places.关键词：kinematic calibration;accuracy compensation;disinfection robot;screw theory;error model212|199|0更新时间：2025-03-15
- 摘要：To enhance the path tracking performance of unmanned agricultural vehicles in complex working scenarios with multiple obstacles, this paper proposes an anti-disturbance predictive control scheme based on safe distance. This control scheme leverages the autonomous learning capabilities of predictive control to achieve efficient obstacle avoidance maneuvers for unmanned agricultural vehicles in complex operational scenarios through interaction with the working environment. Initially, an extended state observer is designed to accurately estimate the unknown disturbance within the agricultural vehicle system and incorporate it into the nonlinear predictive model, thereby improving the precision of state prediction and disturbance rejection in the path tracking control system. Subsequently, an automatic optimization algorithm for the reference point of the agricultural vehicle is designed to prevent excessive steering during path tracking. Then, a novel obstacle avoidance penalty term is constructed based on logarithmic function scaling to ensure that the agricultural vehicle maintains a preset safe distance from multiple obstacles during path tracking. Building on this, the cost function is integrated with lateral offset and heading offset information, combined with the obstacle avoidance penalty function, and an anti-disturbance model predictive control scheme is constructed by solving the nonlinear constrained optimization problem online. Simulation results demonstrate that the control scheme proposed in this paper has superior path tracking accuracy and can effectively avoid obstacles while adapting to different safe distance settings and obstacle positions.关键词：unmanned agricultural vehicle;path tracking control;model predictive control;obstacle avoidance172|187|0更新时间：2025-03-15
- 摘要：Port environments differ significantly from conventional driving environments, characterized by high vehicle activity, customized roadways, and mixed traffic involving human-operated and autonomous vehicles. To address the lack of port-specific autonomous driving data and improve algorithm generalization while reducing development costs and time, this paper proposes a simulation environment for port autonomous driving called DK-Port. Leveraging embodied intelligence principles, DK-Port provides a realistic and controllable environment for vehicle interactions. This paper incorporates survey-based expert input, employing zero-shot prompting and chain-of-thought reasoning techniques to enable large language models to assist in designing reward functions efficiently. Then, human-machine mixed-traffic scenarios are constructed using complex road data, and adversarial driver models are trained with the PPO reinforcement learning algorithm to identify safety vulnerabilities in autonomous driving systems. Comparative experiments in four typical scenarios, including straight roads and intersections, demonstrate that DK-Port generates diverse adversarial driving behaviors aligned with real port characteristics, such as dangerous overtaking and abrupt lane cutting. Under the premise of ensuring a reasonable distribution of key metrics, the number of lane changes in the straight road scenario is 2.9 times that of the baseline method, and the emergency braking rate in the intersection scenario is increased by 46.7%.关键词：ports autonomous driving;simulation environment construction;large language model;reinforcement learning;adversarial scenario generation189|217|1更新时间：2025-03-15
- 摘要：As a special type of robot, tilt-rotor unmanned aerial vehicles have demonstrated great potential in performing tasks in complex environments. To improve its autonomy and adaptability to complex environments, robust flight control law and control allocation method was studied. A cascaded flight control framework based on two-level control allocation was proposed for the significant changes in control strategy and control effectiveness of tilt-rotor unmanned aerial vehicle in different flight modes. A top-level controller was developed for various control modes, and the actual control quantity was generated by a torque control allocation module in conjunction with attitude angle commands using a force control allocation module based on the weighted generalized inverse method. The attitude controller was designed based on a dual closed-loop control structure, and a disturbance observer was added to improve the robustness to external disturbances. Finally, the effectiveness and robustness of the flight control framework and controller were verified by full process simulation and transition simulation under interference conditions.关键词：flight control;tilt-rotor unmanned aerial vehicle;control allocation;weighted generalized inverse method487|585|0更新时间：2025-03-15
- 摘要：Aquatic organisms can sense the flow state and adjust the attack angle in time, thereby maintaining the highest movement efficiency during the periodic flapping process. The biological caudal fin, accompanied by changes in tail stiffness, depends on the contraction/relaxation of tail muscles. In the interaction with external fluid, attack angle adjustment helps organisms maintain the optimal attack angle during the movement to improve propulsion performance. Inspired by this, an intelligent control method for high-speed movement of bionic robots based on attack angle feedback was proposed. Firstly, the kinematic characteristics of the designed robotic dolphin was analyzed to clarify the hardware basis of the algorithm. Secondly, in view of the problem that the caudal fin motor cannot effectively cooperate with the lumbar joint motor during high-frequency movement, a joint angle control algorithm based on fuzzy reasoning was proposed to realize the tracking of the expected joint angle by the caudal fin joint in the torque mode. In addition, a closed-loop control scheme for the caudal joint based on attack angle feedback was proposed to improve movement performance. Finally, the experiments verify the effectiveness of the proposed motion control scheme.关键词：energy efficiency optimization;angle of attack feedback;intelligent control;bionic robot167|156|0更新时间：2025-03-15
Special Column: Embodied Intelligence and Specialized Robot
- Postal code：100079
- Tel：（010）53879206 Email：cjist@bjxintong.com.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
- Total Visits：161982 Visits Today：145
0