张雯

个人信息

参与实验室科研项目

人机混合智能系统双层智能测试评估技术研究

复杂环境下非完全信息博弈决策的智能基础模型研究

学术成果

共撰写/参与撰写专利 1 项，录用/发表论文 2 篇，投出待录用论文0篇。 联培学生可能有其他不在此展示的论文/专利。

patent

1. 一种基于通信时延感知的远程驾驶风险预警方法 李鹏飞, 王若山, 赵云波, 刘金伟, 张雯, and 黄康杰 2025 [Abs] [pdf]

   本发明公开了一种基于通信时延感知的远程驾驶风险预警方法，首先收集目标通信链路的相关数据，对数据进行预处理和特征提取，基于极端梯度提升XGboost算法拟合时延与多特征的非线性关系，实时预测时延的变化情况；构建通信风险评价模型，基于时延变化情况量化出通信风险值；根据车辆动力学模型和驾驶员执行指令序列，预测车辆轨迹；基于车辆轨迹，计算车道偏离风险和障碍物碰撞风险，环境风险值是两者之和；综合考虑环境风险值和通信风险值，根据设计的安全阈值判断是否触发警报。该方法在考虑通信时延的基础上重新设计环境风险评价方式，将环境风险与通信风险综合纳入预警框架，从而为远程驾驶系统提供更好的安全保障。

Journal Articles

1. Multi-Dimensional Shared Control and Its Application to Human-Machine Shared Driving Wen Zhang, Yun-Sheng Zhao, Pengfei Li, Yun-Bo Zhao , and Yu Kang Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 2026 [Abs] [doi] [pdf]

   Shared control is a collaborative paradigm that integrates the strengths of human and intelligent machine. With the advancement of technology, there is an increasing demand for more sophisticated control of dynamic systems, leading to increasingly need for multi-dimensional control. Current research remains predominantly confined to single-dimensional frameworks, proving inadequate for addressing requirements and potentially leading to performance degradation or even task failure. Simply extending single-dimensional cannot resolve this issue, as it requires collaboration among multiple control variables. To address these challenges, this paper proposes a multi-dimensional authority allocation strategy based on dynamic authority boundaries. First, a method for constructing dynamic authority boundaries is proposed for factors that significantly affect shared control performance. Second, a multi-dimensional authority optimization problem is designed on the basis of dynamic authority boundaries to determine the optimal authority allocation. Finally, the specific application of the proposed dynamic authority boundaries and multi-dimensional authority allocation strategy is demonstrated in a human-machine shared driving scenario, and experiments conducted on a human-in-the-loop test platform validate the effectiveness of the framework. Moreover, the proposed framework is not only applicable to shared driving but can also be extended to other domains such as shared control mobile robots and shared control aerial vehicles.

Conference Articles

1. A Shared Control Strategy Considering Control Fusion Security for Human-Machine Co-Driving System Xiaojun Zhu, Bin Lan, Wen Zhang, Yun-Bo Zhao , Yu Kang, and Binkun Liu In 2025 44th Chinese Control Conference (CCC) 2025 [Abs] [doi] [pdf]

   In human- machine cooperative control, it should be ensured that the system is safe and stable as well as efficient in accomplishing the task. Especially in safety critical systems, such as human-machine co-driving systems, ensuring safety is the first priority and other performances may need to be sacrificed to ensure safety. In this case, it is difficult to guarantee other performances of the human-machine co-driving system (e.g., tracking performance). Existing shared control methods cannot guarantee good tracking performance while keeping the system safe. To address this problem, this paper proposes a shared control strategy that takes security into account to ensure system security and stability while realizing good trajectory tracking performance. The effectiveness of the proposed strategy is verified in experiments. The results show that the proposed shared control strategy reduces the trajectory tracking error while ensuring safety.

博客文章