教师名录
李衍杰
通讯地址:深圳市南山区西丽深圳大学城哈工大校区D202O
电子邮件:autolyj@hit.edu.cn; lyj@hitsz.edu.cn
联系电话:

研究方向

随机决策与优化;离散事件动态系统;强化学习;无人机控制;复杂系统控制与优化.

教育经历

2001-2006  中国科学技术大学 自动化系 硕/博士
1997-2001    青岛大学 数学系 学士

研究与工作经历

2010-至今  哈尔滨工业大学 深圳研究生院 副教授
2008-2010  哈尔滨工业大学 深圳研究生院 助理教授
2006-2008  香港科技大学 电子与计算机工程系 博士后
2013   University of New South Wales, Visiting Fellow

专业资质与学术兼职

2010-至今  IEEE 会员
2012-至今  中国运筹学会会员

科研项目

2011-2013  国家自然科学基金:半Markov决策过程基于灵敏度的优化及应用(61004036) 主持
2011-2012  教育部博士点新教师基金:基于性能灵敏度的平均报酬强化学习方法研究(20102302120071) 主持
2011-2013  深圳市基础计划:基于多机器人的智能仓储系统关键技术(JC201005260179A) 主持
2011-2013  国家自然科学基金:基于机器人网络的多机器人同步定位与制图研究 (61005063) 主要参与
2012-2014  深圳市基础计划重点:智能电网输电线路无人直升机检测的关键问题研究(JC201104210048A) 副主持

科研成果及奖励

2014  何潘清漪优秀论文奖
2013  深圳市“孔雀计划”海外高层次人才计划
2006  中国科学院院长奖

发明专利

1. 一种动态电源管理方法,2014100736388.

论文及著作

[1] Yanjie Li, Xinyu Wu, A unified approach to time aggregated Markov decision processes, Automatica, 67:77-84,2016. (SCI)
[2] Yanjie Li, Fang Cao,A basic formula for performance gradient estimation of Semi-Markov decision processes, European Journal of Operational Research, vol. 224, 333-339, 2013. (SCI)
[3] Yanjie Li, Baoqin Yin and Hongsheng Xi, Finding optimal memoryless policy of POMDPs under the expected average reward criterion, European Journal of Operational Research, vol. 211, 556-567, 2011. (SCI)
[4] Yanjie Li, Fang Cao and Xiren Cao, On-line policy gradient estimation with multi-step sampling, Discrete Event Dynamic Systems, vol. 20, 3-17, 2010. (SCI)
[5] Yanjie Li, Baoqun Yin and Hongsheng Xi, Partially observable Markov decision processes and performance sensitivity analysis, IEEE Transactions on System, Man and Cybernetics, Part B, vol. 38, no. 6, 1645-1651, 2008. (SCI)
[6] Baoqun Yin, Yanjie Li, Yaping Zhou and Hongsheng Xi, Performance optimization of semi Markov decision processes with discounted cost criteria, European Journal of Control, vol. 3, pp. 1-10, 2008. (SCI)
[7] Baoqun Yin, Guiping Dai, Yanjie Li, and Hongsheng Xi, Sensitivity analysis and estimates of the performance for M/G/1 queuing systems , Performance Evaluation, vol. 64, no. 4, pp. 347-356, 2007. (SCI)
[8] Guiping Dai, Baoqun Yin, Yanjie Li and Hongsheng Xi, Performance optimization algorithms based on potential for semi Markov control processes, International Journal of Control, vol. 78, no. 11, pp. 801-812, 2005. (SCI)
[9] 唐波,李衍杰,殷保群,连续时间部分可观Markov决策过程的策略梯度估计,控制理论与应用,2009,26(7):805-808.
[10] 殷保群,李衍杰,周亚平,奚宏生,可数半Markov决策过程折扣代价性能优化,控制与决策,2006,21 (8):933-936.
[11] 殷保群,李衍杰,唐昊,半Markov决策过程折扣模型与平均模型之间的关系,控制理论与应用,2006,23 (1):65-68.
[12] 代桂平,殷保群,李衍杰,奚宏生,半Markov控制过程基于性能势仿真的并行优化算法,中国科学技术大学学报,2006, 36(2): 183-186.
[13] 殷保群,李衍杰,奚宏生,周亚平,一类可数Markov控制过程的最优平稳策略,控制理论与应用,2005,22 (1):43-46.
[14] 殷保群,李衍杰,周亚平,奚宏生, 半Markov控制过程在折扣代价性能准则下的最优性方程,控制与决策,2004,19(6):691-694.
[15] 李衍杰,殷保群,奚宏生,周亚平,代桂平, 半Markov过程基于性能势的灵敏度分析和性能优化,控制理论与应用,2004,21(6):1032-1035.
[16] 李衍杰,殷保群,奚宏生,代桂平,一类连续时间Markov链在折扣准则下的灵敏度分析和性能优化,中国科学技术大学学报,2004,34(6):704-709.
[17] 秦廷,陈宗海,李衍杰,递推最小二乘算法的补充性证明,系统仿真学报,2004,16(10):2519-2164.

会议论文及发表演说

[1]Zhang Ruru, Li Yanjie, Lou Yunjiang, Convex optimization of battery energy storage station in a micro-grid, IEEE International Conference on Information and Automation, 2013.
[2]Zeng Wenwu, Zhu Xiaorui,Li Yanjie,Li Lei, Performance analysis of a small-scale unmanned helicopter under large wind disturbance, Proceedings of the 32nd Chinese Control Conference, Xian, 2013.
[3]Yanjie Li, An average reward performance potential estimation with geometric variance reduction, The 31th Chinese Control Conference, Hefei, 2012.
[4] Yanjie Li, Sensitivity-based optimization of semi-Markov decision processes, INFORMS International, Beijing, 2012.
[5] Yanjie Li, Reinforcement learning algorithms for semi-Markov decision processes, The 9th IEEE International Conference on Networking Sensing and Control, Beijing, 2012.
[6] Jianjun Li, Jiangong Ren and Yanjie Li, An average-reward reinforcement learning algorithm based on Schweitzer’s transformation, The 31st Chinese Control Conference,2012.
[7] Yanjie Li, Fang Cao, Inifnite horizon gradient estimation for semi-Markov decision processes, 8th Asian Control Conference, Kaohsiung, Taiwan, 2011.
[8] Wenwu Zeng, Xiaorui Zhu, Yanjie Li and Zexiang Li, Less computational unscented Kalman filter for practical state estimation of small scale unmanned helicopters,IEEE International Conference on Robotics and Automation,2011.
[9] Yanjie Li and Fang Cao, An RVI reinforcement learning algorithm for semi-Markov decsion processes with average reward, World Congress on Intelligent Control and Automation, July, 2010.
[10] Qibo Liu, Yi Liu and Yanjie Li ,Combining sub-bands SNR on cochlear model for voice activity detection,International Conference on Asian Language Processing,2010.
[11] Hong Wang, Yanwen Xing, Yanjie Li and Zexiang Li,Dynamic modeling of five-bar manipulator with structurally flexible linkages , World Congress on Intelligent Control and Automation ,2010.
[12] Yanjie Li, Fang Cao and Xiren Cao, An improvement of policy gradient estimation algorithm. In: Lennartson B, Fabian M, Akesson K, Guia A, Kumar R (eds), The Proceedings of Workshop on Discrete Event Systems, Goteborg, Sweden, pp. 168-172, 2008.
[13] Yanjie Li, Baoqun Yin and Hongsheng Xi, The policy gradient estimation of continuous-time hidden Markov decision processes, IEEE International Conference of Information Acquisition, Hong Kong, 2005.
[14] 陶钊榕,陈智超,李衍杰,基于灵敏度的逆向强化学习,第32届中国控制会议,西安,2013.
[15] 段国强,张岳军,李衍杰,朱晓蕊, 四旋翼无人直升机控制算法比较研究,计算机仿真,录用,2013.
[16] 李建军,任建功,李衍杰, 一种基于Schweitzer变换的平均报酬强化学习算法, 第31届中国控制会议,合肥,2012.
[17]李衍杰,殷保群,奚宏生,受约束Markov决策过程基于性能势的优化算法,第24届中国控制会议,广州,2005.
[18]李衍杰,殷保群,奚宏生,连续时间Markov决策过程基于性能势优化,第22届中国控制会议,宜昌,2003.






任教和任导师经历

哈尔滨工业大学深圳研究生院  2007级硕士:郑倩
2008级硕士:任建功,郭宇,王勇,施志华,关昕,刘其波,吴桂远,黄佳德
2009级硕士:段三星,戴正先,徐颖,陈智超
2010级硕士:段国强,窦川川,李建军,李功捷,岳晓娟,贾可辉
2011级硕士:陶钊榕,张岳军,张羽,张如如,李锐锋,张瑞
2012级硕士:刘凤盟,黄涛,叶睿,孙瑞梅,韩利利
任教课程  最优控制(2012年春,2013年春)

主要介绍凸优化理论的相关概念,算法及应用,变分法,最大原理和动态规划相关理论和计算方法。
参考教材:
[1] S. Boyd and L. Vandenberghe, Convex Optimization, NY: Cambridge University Press, 2004.
[2] A. E. Bryson and Y.-C. Ho, Applied Optimal Control, NY: Taylor & Francis, 1975.
[3] D.P. Bertsekas, Dynamic Programming and Optimal Control, Athena Scientific, 2007.

凸优化理论算法及应用(2010年春,2011年春)

主要介绍凸集,凸函数和凸优化等相关概念,对偶理论,几何理解和计算法方法及其在控制,图像处理和通讯等领域的应用。
参考教材:
S. Boyd and L. Vandenberghe, Convex Optimization, NY: Cambridge University Press, 2004.

线性代数(2009年秋,2010年秋,2011年秋)

主要介绍线性空间,商空间,对偶空间,线性映射,行列式与矩阵,欧式空间和谱理论等。
参考教材:
P. D. Lax, Linear Algebra and Its Applications, Second Edition, NJ: John Wiley & Sons, 2007.
  
最后更新:2017-08-07 10:34:51